Metabolic Engineering of Saccharomyces cerevisiae for


THESIS FOR DEGREE OF DOCTOR OF PHILOSPHY
Metabolic Engineering of Saccharomyces
cerevisiae for Sesquiterpene Production
GIONATA SCALCINATI
Systems & Synthetic Biology
Department of Chemical and Biological Engineering
CHALMERS UNIVERSITY OF TECHNOLOGY
Göteborg, Sweden 2012
I
Metabolic Engineering of Saccharomyces cerevisiae for Sesquiterpene Production
GIONATA SCALCINATI
ISBN: 978-91-7385-720-8
© GIONATA SCALCINATI, 2012.
Doktorsavhandlingar vid Chalmers tekniska högskola
Ny serie Nr 3401
ISSN 0346-718X
PhD Thesis
Systems & Synthetic Biology
Department of Chemical and Biological Engineering
Chalmers University of Technology SE-412 96 Göteborg
Sweden
Telephone +46 (0)31-772 1000
Cover: Schematic representation of the integrated metabolic engineering, systems biology,
Synthetic biology and evolutionary engineering approach for the construction of a “yeast cell
factory”
Printed by Chalmers Reproservice
Göteborg, Sweden 2012
II
Dedicated to
My family, the support of my life…
My love, the inspiration of my life…
“Cyclops, you asked my noble name, and I will tell it; but do you give the stranger’s gift, just as
you promised. My name is Nobody. Nobody I am called by mother, father and by all my
comrades”
Odyssey, Chapter 9 line 366
III
Metabolic Engineering of Saccharomyces cerevisiae for
Sesquiterpene Production
GIONATA SCALCINATI
Systems & Synthetic Biology
Department of Chemical and Biological Engineering
Chalmers University of Technology
ABSTRACT
Industrial biotechnology aims to develop robust “microbial cell factories”, to produce an array of
added value chemicals presently dominated by petrochemical processes. The exploitation of an
efficient microbial production as sustainable technology has an important impact for our society.
Sesquiterpenes are a class of natural products with a diverse range of attractive industrial
proprieties. Due to economic difficulties of their production via traditional extraction processes
or chemical synthesis there is interest in developing alternative and cost efficient bio-processes.
Microbial cells engineered for efficient production of plant sesquiterpenes may allow for a
sustainable and scalable production of these compounds. Saccharomyces cerevisiae is one of
the most robust and characterized microbial platforms suitable to be exploited for bioproduction. The hydrocarbon α-santalene is a precursor of sesquiterpenes with relevant
commercial application and was selected as case study. Here, for the first time a S. cerevisiae
strain capable of producing high levels of α-santalene was constructed through a
multidisciplinary system level metabolic engineering approach. First, a minimal engineering
approach was applied to address the feasibility of α-santalene production in S. cerevisiae.
Successively, a rationally designed metabolic control strategy with the aim to dynamically
modulate a key metabolic step to achieve optimal sesquiterpene production was applied,
combined with the engineering of the main regulatory checkpoint of targeted pathway. It was
possible to divert the carbon flux toward the sesquiterpene compound, and the resulting strain
shows a 88-fold improvement in α-santalene productivity. A second round of strain optimization
was performed using a multistep strategy focused to increase precursors and co-factor supply
to manipulate the yeast metabolic network in order to further redirect the carbon toward the
desired product. This approach results in an overall increase of 1.9-fold in α-santalene
productivity. Furthermore, strain improvement was integrated with the development of an
efficient fermentation/ downstream recovery process, resulting in a 1.4-fold improvement in
productivity and a final α-santalene titer of 193 mg l-1. Finally, the substrate utilization range of
the selected platform was expanded to use xylose as alternative carbon source for biorefinery
compatibility, via pathway reconstruction and an evolutionary strategy approach, resulting in a
strain capable of rapid growth and fast xylose consumption. The results obtained illustrate how
the synergistic application of multilevel metabolic engineering and bioprocess engineering can
be used to obtain a significant amount of high value sesquiterpene in yeast. This represents a
starting point toward the construction of a yeast “sesquiterpene production factory” and for the
development of an economically viable bio-based process that has the potential to replace the
current production methods.
Keywords: Metabolic Engineering, Systems Biology, Synthetic Biology, Evolutionary
engineering, Microrefinery, Cell factory, Saccharomyces cerevisiae.
IV
PREFACE
This dissertation represents the tangible results of my PhD study, carried out at the Systems
and Synthetic Biology group (Sys2Bio), Department of Chemical and Biological Engineering,
Chalmers University of Technology in the period between 2008 and 2012, under supervision of
Professor Jens Nielsen. I believe the results obtained in this thesis are just a small drop in a sea
considering the potential applications of the constantly emerging field I had the privilege to work
in during this research period.
When I first came to Chalmers in July 2008 the Department of Chemical and Biological
engineering did not host a Systems and Synthetic Biology group, but every accomplishment
starts with the decision to try, so under the guidance of a phenomenal group leader and
surrounded by a selected group of finest scientist we start from scratch and embrace the
challenge to create what today I consider a group for excellence in systems level metabolic
engineering. In life there’s always an easy way out but I choose the less travelled road; I lost
sight of days, I lost sight of time, I could have been there for hours days or months just figuring
things out, but that did not matter comparing to how exiting and motivating it was and in the
end the hard work paid off.
The title page of this thesis quotes sentences form the ancient Greek poems ΟΔΥΣΣΙΑ (=
Odyssey). My father use to read me the story of the epic voyage of Ulysses (= Odysseus) when I
was a child; just as Ulysses journey the path that brings me to this doctoral dissertation was rich
of uncertain, unforeseen difficulties, overwhelming hurdles, failure, frustration but even joy,
success, happiness, maturation and friendship. Approaching the end of my dissertation, I now
reached my Ithaca and I am holding the hunting bow ready to shoot the arrow through iron axehelve sockets twelve in line to finish this amazing story I thought I dream it only. I do not yet
know what future holds in store for me but I am ready once again to chase my dream…
Gionata Scalcinati
June, 2012
V
LIST OF PUBBLICATIONS
This thesis is based on the following publications & patent.
Patent Application:
I.
Scalcinati G, Knuf C, Schalk M, L. Daviet L, Siewers V, Nielsen J. Modified
microorganisms and use thereof for terpene production. United States
Provisional patent application filed on June 27, 2011 and PCT Patent Application
EP11171612.2 filed on June 28, 2011.
Publications:
I:
Scalcinati G, Knuf C, Partow P, Chen Y, Maury J, Schalk M, Daviet L, Nielsen J,
Siewers V. Dynamic control of gene expression in Saccharomyces
cerevisiae engineered for the production of plant sesquiterpene αsantalene in fed batch mode. Metabolic Engineering. 2012. 14 (2): 91-103.
II:
Scalcinati G, Partow S, Siewers V, Schalk M, Daviet L, Nielsen J. Combined
metabolic engineering of precursors and co-factor supply to increase
α-santalene production by Saccharomyces cerevisiae. Submitted
III:
Scalcinati G and Nielsen J. Optimization of fed batch process for
production of a sesquiterpene biofuel-like precursor α-santalene by
Saccharomyces cerevisiae. Submitted
IV:
Scalcinati G, J.M. Otero JM, Van Vleet J, Jeffries TW, Olsson L, Nielsen J.
Evolutionary engineering of Saccharomyces cerevisiae for efficient
aerobic xylose consumption. FEMS Yeast research, DOI: 10.1111/j.15671364.2012.00808.x.
During this doctoral research additional publications have been co-authored that are not
included in this thesis:
V.
Chen Y, Partow S, Scalcinati G, Siewers V, Nielsen J. Enhancing the copy
number of episomal plasmids in Saccharomyces cervisiae for
improved protein production. FEMS Yeast Research. DOI: 10.1111/j.15671364.2012.00809.x
VI.
Papini M, Nookaew I, Scalcinati G, Siewers V, Nielsen J. Phosphoglycerate
mutase knock-out mutant Saccharomyces cerevisiae: Physiological
investigation and transcriptome analysis. Biotechnology Journal. 2010. 5
(10):1016–1027.
VII.
Hou J, Scalcinati G, Oldiges M, Vemuri GN. Metabolic Impact of Increased
NADH Availability in Saccharomyces cerevisiae. Applied Environmental
Microbiology. 2009. 76 (3): 851–859.
VI
AUTHOR’S 1 PAPER CONTRIBUTION
A summary of the author’s contribution to the publications on which this thesis is based is
provided below:
Paper I
JN, VS and GS designed the study. JN and VS supervised the project. CK and GS performed
the experimental work. SP and JM assisted the molecular biology experiments. YC assisted the
strain physiology experiments. MD and LD assisted the GC/MS analysis of sesquiterpenes. GS
analyzed the data and wrote the manuscript. All the authors discussed the results, edited and
approved the final manuscript.
Paper II
JN and GS designed the study. JN and VS supervised the project. GS performed the
experimental work. SP assisted the molecular biology experiments. MS and LD assisted the
GC/MS analysis of sesquiterpenes. GS analyzed the data and wrote the manuscript. All the
authors discussed the results, edited and approved the final manuscript.
Paper III
JN and GS designed the study. GS performed the experimental work. GS analyzed the data
and wrote the manuscript. JN and GS discussed the results, edited and approved the final
manuscript.
Paper IV
JMO, GS, JVV, JN, LO participated in the design of the study. JMO and GS performed the
experimental work. JMO and GS wrote the manuscript. JVV, TJ, LO, and JN edited the
manuscript. All the authors have read and approved the final manuscript.
1
GS: Gionata Scalcinati; CK: Christoph Knuf; JM: Jerome Maury; JMO: Jose Manuel Otero; JN: Jens Nielsen; JVV:
Jennifer Van Vleet; LD: Laurent Daviet; LO: Lisbeth Olsson; MS: Michael Shalk; SP: Siavash Partow; TJ Thomas
Jeffries; VS: Verena Siewers; YC: Yun Chen.
VII
TABLE OF CONTENTS
Abstract… … … … … … … … … … … … … … … … … … … .… … … … … … … … … … … ........IV
Preface… … … … … … … … … … … … … … … … … … … … … … ..… … … … … … … … … … … V
List of Publications… … … … … … … … … … … … … … … … … … … .… … … … … … ...… ..VI
Author’s Paper contributions… … … .… … … … … … … … … … … … … … ..… .............VII
Table of Content… … … … … … … … … … … … … … … … … … … … … … … … … … … … ..VIII
List of Figures… … … … … … … … … … … … … … … … … … … … … … … … .… … … … … … .X
List of Tables… … … … … … … … … … … … … … … … … … … … … … … … ..… … … ..… … .XI
Abbreviations and Nomenclature… … … … … … … … … … … … … … … … … … ..… … ..XII
CHAPTER
Introduction… … … … … … … … … … … … … … … … … … … … … … … … … … … ...… … … ..1
1.1 Toward a bio-based economy- the rapidly evolving field of industrial biotechnology.......…..1
1.2 Isoprenoids origins and definitions………………………………………………………...….....2
1.3 Market drivers toward microbial production of sesquiterpenes………………………...……..5
1.4 The new era of systems level metabolic engineering-from local to global……………...…….5
1.4.1 Evolutionary engineering…………………………………………………………...…....6
1.4.2 Synthetic biology…………………………………………………………………………7
1.4.3 Systems biology……………………………………………………………………...…..8
CHAPTER 2 Development of a “Microrefinery”… … … … … … … .........................10
2.1 Industrial biotechnology process overview………………………………………………….....10
2.2 Target product of this study sesquiterpene hydrocarbon α-santalene (C15H24)……………..10
2.3 Selection of production host: yeast as suitable platform for sesquiterpene production……11
2.4 Production strategy design……………………………………………………………………...13
2.4.1 Engineering DNA and gene copy number……………………………………….……14
2.4.2 Engineering transcription…………………………………………...…………………..15
2.4.3 Engineering translation-RNA processing……………………………………………...16
2.4.4 Engineering post translation……………………………………………………………17
2.5 Production process design-Industrial microbial fermentation………………………………...18
2.5.1 Batch cultivation………………………………………………………………………...18
2.5.2 Fed-batch cultivation……………………………………………………………………19
2.5.3 Continuous cultivation…………………………………………………………….…….19
2.6. Techno-Economical analysis of sesquiterpene microbial production………………….……20
VIII
CHAPTER 3 Results & Discussion… .… … … … … … … … … … … … … … … … … … ..… 22
3.1 Construction of a yeast “sesquiterpene cell factory”: α-santalene case study…...…………22
3.1.1 Minimal engineering of yeast for sesquiterpene production: expression of
heterologous plant gene in S. cerevisiae……………………………………….………22
3.2 Rationally designed metabolic engineering approach………………………………………...25
3.2.1 Engineering the regulatory checkpoint of the MVA pathway………………………..25
3.2.2 De-regulation of MVA pathway to increase critical precursor pool………….………27
3.2.3 Dynamic control of MVA pathway branch point………………………….…………..27
3.3 Combined metabolic engineering strategy of precursors and cofactor supply for
sesquiterpene production………………………………………………………………………31
3.4 Development of an efficient fermentation and product recovery process……………….…..35
3.4.1 Fed batch in situ product removal (ISPR) integrated bio-process…………………..35
3.4.2 Optimization of ISPR fed-batch process………………………………………………36
3.4.3 Effect of ethanol as alternative carbon source to increase the precursor pool…….38
3.4.5 Double phase chemostat as tool for study metabolically engineered strains……...39
3.5 Intracellular product accumulation and potential derived toxicity…………………………….41
3.6 Expanding substrate utilization range-toward a biorefinery…………………………………..42
CHAPTER 4 Conclusions & Future Prospects… … … … … … … … … … … … … … … ..46
4.1 Conclusions……………………………………………………..…………………………….….46
4.2 Perspectives……………………………………………………………………………………...47
Acknowledgements… … … … … … … … … … … … … … … … … … … … … … … … … … .… .49
References… … … … … … … … … … … … … … … … … … … … … … … … … … … … … … .… 51
Appendix… … … … … … … … … … … … … … … … … … … … … … … … .… … … … … ..… … .60
Paper I
Paper II
Paper III
Paper IV
IX
LIST OF FIGURES
FIGURE 1.1: Different existing biosyntetic routes for isoprenoids production…….……...……….4
FIGURE 1.2: Microbial production timeline for some relevant plant sesquiterpene products…....9
FIGURE 2.1: Key statistics on the natural source of the target compound of this study αsantalene……………………………………………………………………………....10
FIGURE 2.2: Industrial biotechnology process overview……………………………………….....12
FIGURE 2.3: Simplified scheme of the three principal cultivation modes employed during
biotechnological process……………………………………………………………..17
FIGURE 3.1: Plant santalene synthase (SNS) detailed reaction mechanism…………………….23
FIGURE 3.2: Total ion chromatograms, mass spectra and retention times of authentic
standards and bio-produced targets sesquiterpenes compounds………...……..24
FIGURE 3.3: Rationally designed metabolic engineering strategy for overproduce αsantalene………………………………………………………………………...…….26
FIGURE 3.4: Promoter characterization…………………………………………………………....28
FIGURE 3.5: FPP branch point flux distribution in different mutant engineered to overproduce αsantalene………………………………………………………………………………29
FIGURE 3.6: Overview of the multistep genetic engineering approach for increasing α-santalene
production……………………………………………………………………………..33
FIGURE 3.7: Sesquiterpnes productivity in a two phase partitioned glucose limited aerobic
chemostat……………………………………………………………………………..34
FIGURE 3:8: Time course of an aerobic fed-batch culture with exponential sugar feed of S,
cerevisie strains……………………………………………………………………….36
FIGURE 3.9: The configuration of the in situ product removal (ISPR) fed-batch RQ controlled
cultivation process…………………………………………………………………….37
FIGURE 3.10: Development of RQ based feed-control ISPR aerobic glucose limited fed-batch
cultivation……………………………………………………………………………...38
FIGURE 3.11: Set-up of the in situ product removal (ISPR) chemostat cultivation process……39
FIGURE 3.12: Sesquiterpene production performances in a two phases partitioned glucose
limited aerobic chemostat……………………………………………………………41
FIGURE 3.13: extracellular and intracellular sesquiterpenes accumulation profiles during RQ
based double phase fed-batch process…………………………………………….42
FIGURE 3.14: Synthetic pathway reconstruction strategy for xylose assimilation in S.
cerevisiae……………………………………………………………………………...43
FIGURE 3.16: Directed evolution of S. cerevisiae strains for xylose consumption………….......44
FIGURE 3.17: Transcriptome analysis of evolved and unevolved S. cerevisiae strains…………45
FIGURE 4.1: Santalene productivity progression achieved during this study applying different
strategies……………………………………………………………………………....46
X
LIST OF TABLES
TABLE 1.1: Examples of key production platforms of isoprenoids bio-product….……….....…...3
TABLE 2.1: Chemical structure and proprieties of the target compound of this study αsantalene…………………………………………………………………………….…..10
TABLE 2.2: α-santalene maximal theoretical yield and pathway yield under different carbon
sources.……………………………………………………………………................20
TABLE 3.1: Candidate promoter systems description…………………………………………….28
XI
Abbreviations & Nomenclature
NADP+: Nicotinamide adenine dinucleotide
asRNA: antisense RNA
phosphate
B. subtilis: Bacillus subtilis
NADPH: Nicotinamide adenine dinucleotide
phosphate hydrogen
C. lansium: Clausena lansium
C. glutamicum: Corynebacterium glutamicum
CTR: Carbon transfer rate mmol l
-1
D: Dilution rate h-1
Dcrit:: Critical Dilution rate h-1
DNA: Deoxyribonucleic acid
DO: Dissolved oxygen
DXP: 1-deoxyxylulose-5-phosphate
E. coli: Escherichia coli
ER: Endoplasmic reticulum
NPP: Nerolidyl diphosphate
OPP-: Diphosphate anion
OTR: Oxygen transfer rate mmol l-1
Vmax: Maximum reaction rate
Km: Michaelis constant
PERG9: Squalene synthase native promoter
PPP: Pentose phosphate pathway
PUFAs: Polyunsaturated fatty acids
P. stipitis: Pichia stipitis
ERG9: Squalene synthase gene
rasiRNAs: Repeat associated small interfering
RNAs
FPPS: Farnesyl diphosphate synthase
RQ: Respiratory quotient
FPP: (E,E)-Farnesyl diphosphate
Rs: Indian rupee
FOH: (E,E)-Farnesol
S. cerevisiae: Saccharomyces cerevisiae
FAO: Food and Agriculture Organization of the
United Nations
SF: Shake flask
gDCW: Grams dry cell weight of biomass
SanSyn: Santalene synthase gene
GO: Gene ontology
SanSynOpt: Santalene synthase-codon optimized
gene
HMG1: HMG-CoA reductase gene
siRNAs: small interfering RNAs
HMGR: 3-hydroxy-3-metyl-glutaryl-coenzyme A
reductase
SQS: Squalene synthase
LogP: Logarithm (base 10) of partition coefficient
Mb: Mega base; a million of bases
miRNAs: micro RNAs
MVA: Mevalonate
SNS: Santalene synthase
SSD: Sterol sensing domain
$: United States Dollars
TFs: Transcription factors
NAD+: Nicotinamide adenine dinucleotide
tHMG1: Truncated version of HMG-CoA
reductase gene
NADH: Nicotinamide adenine dinucleotide
hydrogen
tHmg1: Truncated version of HMG-CoA
reductase
µmax: Maximum specific growth rate
XII
CHAPTER 1 Introduction
1.1 Toward a bio-based economy- the rapidly evolving field of Industrial biotechnology
Biotechnology is reshaping industrial production, and the past 20 years have witnessed an
exponential increase of bio-based products and bio-energy in the global economy (Enriquez,
2009). The chemical industry is actively searching for alternative routes to petroleum-based
processes influenced by environmental sustainability trends and the need to freeing the
dependency from non-renewable resources. The concept “bio-product” has been known since
the origin of the fermentative solution for production of bread, beer, wine or cheese (Russo et
al., 1995). The movement toward a more green society has driven unprecedented research
focus on the “bio-route” in order to diversifying away from petrochemical feedstock and in an
effort toward a more sustainable development (Otero et al., 2007, Stephanopoulos, 2010).
Industrial biotechnology2 rapidly penetrates in the chemical manufacturing world as concrete
sustainable, renewable and ecologically friendly alternative, allowing developing new biological
products exploiting biological systems, using fermentation technology processes to convert
agricultural basic raw material (e.g. corn syrup) into a wide range of products. The technologies
involved in the industrial biotechnology process are nowadays self evident and sufficiently
mature to reach the final stage of full commercialization. Already in 2005, 7% of chemical sales
depended on biotech, with $77 billion in revenue within the chemical sector (source: McKinsey,
SRI) making industrial biotechnology a realty.
Efficient development in cell factory design is a crucial aspect in the success of industrial
biotechnology. Over the years, tremendous progress has been made to turning biological
systems into “biorefineries3” capable of converting inexpensive raw material into valuable
chemicals. Microbial cellular metabolism has synthetic potential and chemical features that
rarely can be achieved by a chemist under the same physical conditions (e.g. temperature and
pressure). Therefore the field has largely focused on the creation of efficient microbial, self
regenerating, factories to produce chemicals, fuels and material.
Current industrial biotechnology major market segments are represented by specialty chemicals
(31%) base chemicals (25.3%) consumer chemicals (22.5%) and active pharma ingredients
(21.2%) (Festel, 2010). McKinsey & Company forecasted that the global biotech industry
2
Industrial Biotechnology: The application of biotechnology for the processing and production of chemicals,
material and energy (Otero et al., 2007).
3
Biorefinery: Conversion of renewable resources into bio-products (chemicals and materials) and/or energy, via
biocatalysis using microbial fermentation or enzyme catalysis. (Bohlmann 2005; Kamm et al., 2004).
1
revenue has the potential to generate upwards of $300 billion by the year 2020 (McKinsey SRI).
The market driving forces for the biorefineries establishment are attributed mainly to biofuel
(ethanol and biodiesel), however, the projected growth showed how the greatest impact will be
in fine chemicals production (The economist, 2010; Dornburg et al., 2008). In the following, the
use of industrial biotechnology for production of isoprenoids compounds a widespread group of
molecules with a variety of potential applications heavily targeted for biorefinery is examined.
1.2 Isoprenoids origins and definitions
Isoprenoids (often called terpenoids) are a ubiquitous class of natural compounds (over 40,000
different compounds) with many potential commercial applications that have not been fully
explored, e.g. fragrances (linalool, geraniol, menthol etc.), cosmetics (squalane), disinfectants
(camphor, α-pinene), flavoring agents, food colorants (zeaxanthines, astaxanthine), food
supplements (vitamins A, E, K), functional foods (α-humulene), bio-pesticides, nutraceutical and
pharmaceutical agents (taxol, artemisinin). They represent a very diverse class of secondary
metabolites and they satisfy distinct biological functions like pheromones, defensive agents,
photosynthetic pigments, attractants, repellents, toxins, antibiotics, anti-feedants, electron
transporting chain quinones, structural membrane components (McGravey et al., 1995). They
have many different physico-chemical proprieties, lipophilic or hydrophilic, volatile or nonvolatile, cyclic or acyclic, chiral or achiral, reflected in their complexity, due to the multitude of
biological activities they fulfill (Bohmann et al., 2008). They are naturally produced in subsequential head-tail heteropolymeryzation condensation of isoprene functional units, isopentenyl
diphosphate IPP, in all organisms and classified based on the content of isoprene units as:
hemiterpenes (C5), monoterpenes (C10), sesquiterpenes (C15), diterpenes (C20), sesterterpenes
(C25), triterpenes (C30). The isoprene universal building block IPP is naturally synthesized via two
independent pathways: the mevalonate (MVA) pathway and the 1-deoxyxylulose-5-phosphate
(DXP) pathway (Kuzuyama et al., 2003). These two biosynthetic pathways are taxonomically
distributed, the MVA pathway is found in Eukarya, Archaea (a modified version) and a few
bacteria whereas the DXP pathway in Bacteria and photosynthetic Eukarya. Some bacteria and
plants have been shown to have both pathways, and the existence of an alternative MVA
pathway was recently discovered (Lombard et al., 2010) (Fig. 1.1). The MVA pathway starts with
the condensation of three units of acetyl-CoA into the intermediate mevalonate that
successively undergoes phosphorylation and decarboxylation resulting in formation of IPP. The
DXP pathway starts with the production of DXP from pyruvate and glyceraldehyde-3P that is
then rearranged into MEP that reacts with cytidine 5’-triphosphate. The resulting reaction
product is phosphorylated, cyclized and in the final two steps IPP and DMAPP are formed (see
2
Fig. 1.1 for details). The two pathways are compartmentalized differently depending on the
organism and may occur in the cytosol, peroxisome, outer phase of the endoplasmic reticulum
and plastid (Lange et al., 2000).
From the current state of the art, several isoprenoid products are successfully produced or
road-ready and expected to be produced in the near future by a biotech process, and a smallsubset of relevant examples is provided in Table 1.1.
Table 1.1. Examples of key production platforms of isopenoid bio-products
Product
Formula
Company
Application
Source
Isoprene
C 5H 8
Genencore
Rubber
Withed
et
al.,
2010
Artemisinic
acid
C15H22O5
Amyris/Sanofi‐
Aventis
Antimalarial
drugs
precursor
GEN
News,
2008
Farnesene
TM
(Biofene )
C15H24
Amyris/Tate
&Lyle
Biodiesel
GEN
News,
2010
Squalane
C30H62
Soliance/Amyris
Cosmetic
Katie,
2010
In this study, particular focus was dedicated to the sesquiterpenes, a class of compounds
originated from the common precursor farnesyl diphosphate FPP derived from the assembly of
three IPP units (Maury et al. 2005).Sesquiterpenes are one of the largest isoprenoids groups
(over 7000 different compounds) (Misawa, 2011). C15-branched sesquiterpenes are receiving
increasing attention as they may not only serve as precursor chemicals for production of
perfumes and pharmaceuticals but also as precursors for a new generation of biofuels that can
be used as diesel and jet fuels (Peralta-Yahya et al., 2011; Zhang et al., 2011; Rude et al.,
2009; Lee et al., 2008). The portfolio of fuel candidate compounds in fact has been greatly
expanded lately, with special attention dedicated to the drop-in biofuel “class of bio-fuel that
can easily replace gasoline or diesel in existing engines” (Craig et al. 2012), highlighting
branched and cyclic sesquiterpenes as potential jet fuel precursors based on their
physicochemical proprieties (Peralta-Yahya et al., 2011; Renninger et al., 2008).
3
Figure 1.1. Eukaryal mevalonate (MVA) pathway, modified archeal mevalonate (MVA) pathway and bacterial
methylerythritol phosphate (MEP) pathway. (1) glyceraldehyde-3-phosphate, (2) pyruvate, (3) acetyl-CoA, (4)
acetoacetyl-CoA, (5) 3-hydroxy-3-methylglutaryl-CoA, (6) mevalonate, (7) mevalonate-5-phosphate, (8) mevalonate-5diphosphate, (9) isopentenyl pyrophosphate, (10) isopentenyl phosphate, (11) 1-deoxyxylulose-5-phosphate, (12) 2-Cmetyl-D-erythritol-4-phosphate, (13) 4-diphosphocytidyl-2-C-methyl-D-erythritol, (14) 4-diphosphocytidyl-2-C-methylD-erythritol-2-phosphate, (15) 2-C-methylerythritol-2,4-cyclopyrophosphate, (16) 1-hydroxy-2-methyl-2-(E)-butenyl-4pyrophosphate, (17) dimethyallyl diphosphate, (18) geranyl diphosphate, (19) farnesy diphosphate, (20) greanylgeranyl
diphosphate. (ACCT) Acetyl-CoA thiolase, (HMGS) HMG-CoA synthase, (HMGR) HMG-CoA reductase, (MVK)
mevalonate kinase, (PMK) phosphomevalonate kinase, (?) phosphomevalonate decarboxylase (not identified yet), (IPK)
isopentenyl phosphate kinase (MDC) mevalonate pyrophosphate decarboxylase, (IDI) isopentenylpyrophosphate
isomerase, (FPPS) farnesyl diphosphate synthase, (GPPS) geranylgeranyl diphosphate synthase, (DXS) DXP synthase,
(IspC) DXP reductoisomerase, (IspD) 2-C-metyl-D-erythritol-4-phosphate cytidyltransferase, (IspE) 4-diphosphocytidyl2-C-methyl-D-erythritol-2-phosphate kinase, (IspF) 2-C-methylerythritol-2,4-cyclopyrophosphate, synthase, (IspG) 1hydroxy-2-methyl-2-(E)-butenyl-4-pyrophosphate synthase, (IspH) 4-hydroxy-3-methylbut-2-enyl diphosphate
reductase.
4
1.3 Market drivers toward microbial production of sesquiterpenes
As introduced in the previous chapter, the demand for microbial production of chemicals as an
alternative to petrochemical based synthesis is increasing due to economical, environmental
and geopolitical factors (Dellomonaco et al., 2010; Stephanopoulos et al., 2007). Microbial
productions are gaining popularity especially for biosynthesis of added value compounds (Hong
et al., 2012; Kim et al., 2012) due mainly to the small margin achievable from commodity
production. Isoprenoids and isoprene derivative represent nowadays a $650 million global
market (Sims, 2012). Recently, their role as biomaterial resource has been rediscovered leading
to renewed interest in this class of molecules (Bohmann et al., 2008). The complexity of
isoprenoid is often the main drawback for the industrial scale production. Nowadays, most of
the isoprene derived compounds are produced via plant extraction and by total or semisynthesis (Teisserire 1994). Extraction from natural resources is limited by raw material
accessibility, low yields, high process costs and often lead to a complex mixture of products
(Koepp et al., 1995); complete chemical synthesis generally involve multistep transformation
resulting in an inefficient, expensive process and may not result in enantiomeric pure products
(Miyaoka et al., 2002, Mukaiyama et al., 1999, Danishefsky et al., 1996,). The production of
isoprenoids by microbial fermentation is an environmentally friendly and attractive alternative to
the traditional methods and offers several advantages, among them it (i) avoids formation of
racemic mixtures providing pure isomer products through enzymatic biocatalysis; (ii) reduces
process cost using inexpensive sugar based carbon sources, (iii) increases sustainability
avoiding harvesting and extraction from natural sources and thus reducing environmental
footprint, lowering CO2 emissions and toxic waste e.g. solvents and metal catalysts (iv)
increases yield and productivities using genetic manipulation of the heterologous host and (v) is
compatible with scalable high density fermentation processes. This has caused interest in
engineering cell factories that can be used to produce isoprenoids in a cost competitive fashion
(Khalil et al., 2010; Koffas et al., 2009; Fortman et al., 2008).
1.4 The new era of systems level metabolic engineering-from local to global
Metabolic engineering2 is a constantly evolving field and has driven for years the construction of
recombinant microorganism for the production of target compounds. Metabolic engineers have
relied for long time on traditional and intuitive approaches to bioengineer microbial cells to
produce desired chemicals. However, through the years it appears clear that the hierarchical
complexity of cell regulation requires a systems level approach moving from local to global
applications. The need of and holistic access to the cellular network leads to the synergistic
application of related emerging disciplines: systems biology, synthetic biology and evolutionary
5
engineering (Box 1.1) opening new opportunity for cellular engineering and creating the
intertwining that produced the modern multi-disciplinary field of metabolic engineering (Nielsen
et al., 2012, Lee et al., 2011a). The integration and impact of these different disciplines for
metabolic engineering is briefly introduced in the following, with the techniques mostly applied
through this research study being addressed.
Box.
1.1.
1.4.1 Evolutionary engineering1
Evolutionary
engineering:
The
application
of
a
selection
procedure
to
obtain
a
desired
phenotype
¥.
2
Metabolic
engineering:
Use
of
genetic
engineering
modifications
to
manipulate
cell
factories
with
the
objective
to
improve
their
proprieties
for
industrial
application
‡.
3
Synthetic
Biology:
Design
and
construction
of
new
biological
components,
functions,
and
genetic
circuits
de
novo
or
redesign
existing
biological
systems
†.
4
Systems
Biology:
To
obtain
new
insight
into
the
molecular
mechanism
occurring
in
living
cells
for
predicting
the
function
of
biological
systems
through
the
combination
of
mathematical
modeling
and
experimental
biology
#.
Evolutionary approaches have been widely used
‡
Sources:
Bailey
et
al.,
1991
&
Stephanopoulos
#
et
al.,
1991;
Nielsen
et
al.,
2001;
†Keasling
et
¥
al.,
2008;
Sauer
et
al.,
2001
alternative to the classical mutagenesis technique.
1
to improve the properties of industrial cell
factories:
the
creation
of
novel
metabolic
functions, expanding substrate utilization range,
improve the growth rate, improve tolerance
towards
multiple
compounds,
biocatalysis
and
phenotypes
(Cakar
evolutionary
procedures
many
et
based
on
other
al.,
methods
improve
2010).
refer
the
favorable
to
use
Directed
selection
of
specific
environmental pressures through iterative genetic
diversification
with
the
final
goal
of
strain
improvement (Chatterjee et al., 2006). These
methods
exploit
natural
rationally
applied
and
Among
the
existing
selective
offer
a
multitude
pressure
non-invasive
of
adaptive
evolutionary approaches the most popular are (i)
extended chemostat cultivation (Jensen et al.,
2005; Sauer et al., 2001;) and (ii) repetitive batch cultivation (Barrick et al., 2009; Kuyper et al.,
2005), performed under selective conditions. Evolutionary engineering has been frequently
combined with metabolic engineering from the early days of industrial biotechnology as simple
methods to overcome cellular complexity because of the capacity to address multi-gene traits
(e.g. resistance to toxic compounds) that can be difficult to solve with rational approaches. The
common limitation of this approach is the dependency on the screening method and the
random outcome and the inability to elucidate the mechanisms that confer the adaptive fitness.
However, recent advances in high-throughput techniques and DNA sequencing efforts have
facilitated the identification of genetic modifications driving identified phenotypes and hereby
6
greatly enhanced the application of this technique In this study, evolutionary engineering was
applied to expand the spectrum of usable carbon sources of the selected cell factory in order to
open the possibility to efficiently use alternative feedstocks like lignocellulose as raw material
(Ritter, 2008). Due to its global abundance and renewability lignocellulose is an attractive
starting material for bio-production of value added products (Chapter 3.5).
1.4.2 Synthetic Biology
Synthetic biology2 can be envisaged as the extension of engineering principles to genetic
engineering by biologists involving the design/redesign of devices and circuits for controlling
biological systems (Endy, 2005). The impact of synthetic biology on metabolic engineering is
rapidly reshaping the industrial biotechnology field (Keasling, 2012). The dramatic decrease in
the cost of whole genome sequencing and long-chain DNA synthesis has led to the
development of modern synthetic biology tools and methodology bringing new prospects and
un-restricted access to microbial pathway engineering (Smolke et al., 2012, May, 2009).
Synthetic biology has influenced the bioresearch field by making cell factory development faster
and more efficient allowing wider exploration of the biosynthetic potential of microbial
production and advancing our metabolic engineering capabilities (Keasling, 2010). The diverse
set of tools emerged for pathway engineering increase the capability to achieve specific cellular
functions (Canton et al., 2008). It is generally accepted that pathway engineering requires a
balanced expression of single and multiple genes avoiding wasteful and potentially toxic
intermediate accumulation and preventing “robbing” of the cell of key precursors. Additionally,
traditional overexpression technique may result in high protein levels resulting in unwanted
metabolic burden. Therefore, an optimization strategy should be carefully designed, and
synthetic biology can be used to introduce synthetic sensors like dynamic control element able
to sense cellular metabolic state and regulate the expression of specific functions (Farmer et al.,
2000, Zhang et al., 2011) and hereby shed light on the importance of the dynamic aspect of
pathway engineering (Holtz et al., 2010).
In this study, a synthetic biology concept was applied combining a static engineering module
with dynamic control for pathway engineering. Remodeling of the cellular network was
conducted using an environment–responsive promoter to dynamically control the gene
expression of a regulatory branch point in response to an extracellular signal molecule
concentration and modulating the flux between the target pathway and three branches (see
Chapter 3.2.3). An attempt to create a dynamic driving force along the engineered pathway was
performed modifying cellular cofactor availability (see Chapter 3.5). In this work, a synthetic
7
pathway for expanding substrate range capability was also re-constructed in the production
host (see Chapter 3.6).
1.4.3 Systems Biology
Systems biology3 aims to get insight into the complexity of cellular functions offering the
opportunity to understand and optimize cellular processes through the combined use of highthroughput experimental methods (top-down approach) and computational models (bottom-up
approach). The ability to obtain a quantitative analysis of the whole cellular system is
strategically useful during the design of a novel cell factory (Nielsen et al., 2007). Advances in
high-throughput technique allow rapid cellular phenotype characterization affecting the ability to
engineer cell metabolism. The systems biology toolkits (x-omics) routinely applied for this
purpose include: genomics, transcriptomics, metabolomics, fluxomics (Petranovic et al., 2009).
On the other hand, the availability of detailed mathematical models expands analytical access to
strain engineering; the predictive capacity of in silico analysis of metabolic flux distribution is
crucial in guiding the strain improvement identifying potential targets for modification required to
achieve desired performances (Patil et al., 2004 Stephanopoulos et al., 1999). Moreover, the
capability of exploring multiple possible flux distribution scenarios using computational analysis
saves time and costs required for in vivo experimentation, selecting the best set of modifications
out of large number of potential combinatorial changes and further delineating strain
construction strategies (Burgard et al., 2003; Patil et al., 2005). Sophistication in bioinformatics
for system level data handling greatly contribute to the integration of the different “x-omics”
dataset enhancing the application of this techniques and changing the way in which metabolic
engineering is executed.
In this study, systems biology was applied at two levels: (i) Transcriptome analysis, one of the
most developed and implemented “x-omics” tools for metabolic engineering (Jewett et al.,
2005), was employed to further elucidate metabolism and physiology of the mutant obtained
through evolutionary techniques (see Chapter 3.5); (ii) A non-intuitive systematic strategy
obtained from previously performed in silico analysis using a genome scale metabolic model
(Asadollahi et al., 2009) was applied to manipulate the cellular cofactor balance of the
constructed cell factory in an attempt to empower flux toward the target product (see Chapter
3.3).
Although the above mentioned disciplines are quite different the high level of interconnection
allows their simultaneous application for bioengineering purposes. In the past decade,
multidisciplinary system level metabolic engineering approaches have started to have a strong
8
impact in the biological production of sesquiterpene derived compounds and the number of
reports of engineered microorganisms producing sesquiterpene compounds has risen
dramatically making the microbial production of these series of compounds an industrial reality
(Fig 1.2).
Figure 1.2. Microbial production timeline for some relevant plant sesquiterpene products. Synthetic biology
advanced the classic metabolic engineering approach leading to dramatic improvement in final titers achievable.
The list of examples provided is by no means exhaustive and it is intended to provide an overview of the context
referred. Reference data, Martin et al., 2001; Jackson et al., 2003; Martin et al., 2003; Ro et al., 2006; Takahashi
et al., 2007; Asadollahi et al., 2008; Wang et al., 2011a; Albertsen et al., 2011; Peralta-Yahya et al., 2011;
Westfall et al., 2012.
1
6
2
7
3
8
4
5
9
10
Today, the creation of “superbugs” requires a dynamic interaction and application of all these
disciplines (Nielsen et al., 2011). Among several successful examples of how this combined
approach has impacted industrial biotechnology the yeast-based production of the anti-malaria
drug precursors amorpha-4,11-diene and artemisinic acid represent a remarkable achievement
(Westfall et al., 2011) (Fig 1.2). Another salient example is the bacterial production of taxol
precursors taxadiene and taxadien-5α-ol (Ajikumar et al., 2010).
9
CHAPTER 2 Development of a “M icrorefinery 4 ”
2.1 Industrial Biotechnology process overview
Development of a biotechnological process involves different phases (i) target product
identification (ii) selection of a suitable production host (iii)
production strategy design and (iv) production process
design, including the cost and accessibility of the raw
material (e.g. the carbon source) (Fig. 2.2). During the early
design stage it is important to take into consideration the
entire process and integrate together the different steps
avoiding pitfalls moving from one stage to another. Typically,
process optimization proceeds via several rounds of cyclic
optimization. The result of the metabolic engineering efforts
are evaluated by available screening techniques, bottlenecks
are being identified and another round of optimization takes
place.
2.2.
Target
product
of
this
study,
sesquiterpene
hydrocarbon α-santalene (C15H24)
Natural products are the most valuable fragrances, but limited
access to many of these compounds has led the perfume
Figure 2.1 Key statistics on the natural
source of the target compound of this
study
α-santalene.
*Adapted
from
Essential Oils the new crop industries
handbook RIRDC 2004 (Rs=17$)
α-(+)-santalene industry to look for artificial substitutes (Chapuis et al., 2004).
The woody fragrance sandalwood for examples is one of the
most expensive perfumery raw materials and its components
Molecular
Formula
are extremely difficult to synthesize (Davies 2009). α-
Molar
Mass(g
mol )
Santalene (CAS Number: 512-61-8; IUPAC Name: [(-)-1,7dimethyl-7-(4-methyl-3-pentenyl)-tricyclo
(2.2.1.0
(2,6))
heptane]) (Table 2.2) is the precursor of the hydroxylated αsantalol one of the main components of the East Indian
saldalwood oil (Corey, 1957; Baldovini, 2010). The extracted
essential oil is among the most precious and highly prized
world’s fragrances. α-Santalol together with ß-santalol are
‐1
‐1
Density
(Kg
l )
Molar
energy
density
‐1
(kJ
mol )
Mass
energy
density
‐1
(MJ
Kg )
Boiling
point
(°C
at
760
mmHg)
C15H24
204.35
0.944
9148
44.8
247.6
Table 2.2 Chemical structure and
properties of the target compound of this
study, α-santalene.
4
Microrefinery: Microbial system conversion of renewable resources into bio-products chemicals and fuel (LS9,
INC).
10
the main olfactory components of the sandalwood oil that can contain up to 90% of this
sesquiterpene alcohol (60-50%-α, 30-20%-ß) and confer the sweet-woody, warm, animal and
milky-nutty scent employed for centuries in religious and cultural contexts (Howes et al., 2004;
Schalk, 2011; Brunke et al., 1995). Sandalwood essential oil is mainly extracted from tree and
roots of the two plant species Indian sandalwood (Santalum album) and Australian sandalwood
(Santalum spicatum). In the past decade, the sandalwood oil price has skyrocketed due to
intensive harvesting that rendered the Indian tree an endangered species and governmentally
protected (FAO 1995) and the constant increase in demand (Fig 2.1). India is the major supplier
of sandalwood oil, but the international scenario is quickly changing (Misra 2009). Nowadays,
the market price is estimated to lie between $1.200-2.700/ kg depending on the quality
(http://www.alibaba.com/). However, because the content of α/ß santal-ol/ene determines the
oil market price (Nautiyal 2011), the 100% pure santalene α-(+) isomer price could be up to 10
fold higher. Besides its commercial use in cosmetic, perfumery and aromatherapy industries
sandalwood oil finds application as chemotherapeutic and chemopreventing agent against skin
cancer (Dwivedi et al., 2003) and for its antimicrobial (Jirovetz et al., 2006) and antiviral
proprieties (Benecia et al., 1999).
2.3 Selection of production host: yeast as suitable platform for sesquiterpene production
The choice of microbial host is dictated by many factors and often requires a trade-off; here are
discussed some of the aspects that need to be considered in order to fulfill the industrial
demands. Among desirable features of the selected microorganism are (i) the metabolic
capability toward the desired product; (ii) high substrate utilization rate and ability to grow fast
on minimally supplemented media and synthesize all the required macromolecules for growth
from inexpensive C source and N, P, S salts avoiding the supplement of complex nutrients; (iii)
tolerance to inhibitory compounds potentially present in the industrial fermentation media (e.g.
hydrolyzate tolerance) or intermediate metabolites and side products produced along the
process; (iv) robustness toward the target compound itself; if the selected organism can tolerate
a certain concentration of final product this limit cannot be exceeded without resulting in toxic
effects; (v) resistance to adverse environmental conditions; ideally the suitable host should
tolerate elevated temperatures (thermo-tolerance), low pH (acid-tolerance) and high osmotic
pressure (osmo-tolerance) reducing cooling costs, probability of contamination and osmotic
pressure derived from elevated concentrations of nutrients or products; (vi) capacity to efficiently
perform regardless of environmental changes during the production process; (vii) genetic
tractability, considering capacity to integrate and efficiently express heterologous DNA and high
transformation efficiency; (viii) genetic stability during extended cultivation periods; (iv) the
11
availability of metabolic engineering tools and (x) genome wide characterization including access
to the “x-omics” analysis tools.
Figure 2.2. Industrial biotechnology process overview. The first step consists in the identification of the compound
to be produced and the selection of a suitable production host. Second, a production strategy design including
genetic, enzyme and biosynthetic pathway engineering is developed. Third, fermentation and downstream process
are performed to produce the final target. Process efficiency is obtained through several cycles of optimization of the
different steps proposed.
Considering the number of variables involved host choice is clearly not one solution problem.
Often the decision lies between engineering recombinant microorganisms or exploring the
potential of native producer microorganism (Alper et al., 2009). Depending on the target
compound non-recombinant microorganisms may have high process capability and a high level
of toxicity resistance but the lack of tools and detailed physiology knowledge could require
costly and time demanding research efforts in order to establish an efficient process.
Traditionally applied “model organisms” (e.g. E. coli, S. cerevisiae, A. niger, B. subtilis, C.
12
glutamicum) are on the other hand well characterized and easy to manipulate but they might
lack the required industrial robustness. The sophistication of systems and synthetic biology
tools have largely improved the capacity to manipulate model microorganisms and accelerate
the process to achieve efficient “microrefineries” expanding their potential of model organism
and making them more attractive platforms (Enyeart et al., 2011) (see Chapter 1.4.2 & Chapter
1.4.3 for details). In this study the S. cerevisiae laboratory strain CEN.PK113-7D, which is widely
applied for industrial biotechnology applications (van Dijken et al., 2000), was selected as
starting point for the development of sesquiterpene bio-production. For S. cerevisiae, there are
well-characterized genetic manipulation protocols, detailed physiology records, advanced
metabolic engineering tool set to perform precise gene expression, it has been extensively
characterized with high-throughput approaches (genomics, transcriptomics, proteomics,
metabolomics, fluxomics); computational methods (e.g. genome scale models) are available for
guiding in silico experimental design and data analysis. It has a generally regarded as safe
(GRAS) status and has been widely applied in successful industrial processes. S. cerevisiae was
identified as best ergosterol producer among over 69 yeast species (Dulaney et al., 1954), and
the CEN.PK background strains displays a high ergosterol content during growth on glucose
(Daum et al., 1999). Ergosterol is produced in yeast through the sterol pathway from the final
product of the MVA pathway, FPP, from which sesquiterpenes are also derived (see Chapter
1.2 & Fig 1.1). Recently, the whole genome sequence of CEN.PK113-7D was completed and
SNP analysis revealed that the strain specific high sterol biosynthetic capacity may be due to
genetic variation of several genes in the MVA pathway (ERG8, ERG9, ERG12, HMG1 and
ERG20) compared to the reference strain with lower ergosterol content (Otero et al., 2010).
Additionally, greater variability was found in the promoter region of the same genes
(http://www.sysbio.se/cenpk). The combination of all these characteristics can be capitalized
upon to enable the industrial application of S. cerevisiae CEN.PK113-7D as production host
and favored the choice as production host in this study.
2.4 Production strategy design-Pathway engineering
Once product and host have been selected a production strategy needs to be designed.
Typically, production strategy optimization is an iterative process where the simultaneous
regulation and timing of the expression of multiple heterologous and native genes is required for
the redirection of the metabolic flux towards the target compound. Common pathway
engineering operations include (i) re-engineering of existing pathways, (ii) combination of existing
pathways with exogenous or synthetic novel functions, (iii) de novo assembly of new pathways.
13
An important aspect concerns the optimization of endogenous pathways compared to the
import of heterologous functions (Alper et al., 2009). Target compounds of this study,
sesquiterpene derivatives are naturally produced by S. cerevisiae network thus, through
engineering strategy was focused on optimize the yeast native MVA pathway. Different
approaches were performed aimed to maximize the flux through the MVA pathway, increase the
flux to the MVA pathway and redirect the flux at the branch point of the MVA pathway. On the
other hand, a “bioprospecting5” approach was used to import a novel pathway to expand the
substrate capability of the designed cell factory.
Cellular networks have evolved the ability to rapidly sense and respond to environmental
changes. When perturbations are introduced in an attempt to increase flux toward a desired
path there is a risk to produce unexpected and unwanted responses as a result of flux
imbalance that could result in host instability. Metabolic engineering side effects that limit the
final yield can be ascribed to (i) poor understanding of the complex cellular regulation; (ii)
unbalanced consumption of cellular resources (e.g. cofactors imbalance, precursors pools) (iii)
metabolic burden of heterologous protein; (iv) accumulation of toxic intermediate, (v) toxicity or
inhibitory effect of the final product, metabolites and heterologous enzymes, (vi) negative
feedback loop; (v) poor expression of desired new component.
Therefore, a number of tools have been developed to control and coordinate fluxes through
different branches in the cellular network such that there is maintained a balance between the
resources required for cell growth and the precursors for target compounds. Engineering of
biological systems can be realized at multiple levels: gene number, transcription, posttranscription, post-translation (Young et al., 2010; Boyle et al., 2009; Nevoigt et al., 2008).
2.4.1 Engineering DNA and gene copy number
The DNA-level manipulation toolset for pathway engineering comprises plasmids vectors, and
chromosomal integrations methods (Siddiqui et al., 2012; Siewers et al., 2010). Plasmid vectors
are the most common and widely applied gene expression tools for metabolic engineering.
Recently, commercial cloning vectors available for yeast use have been reviewed in detail (Da
Silva et al., 2012). Among the desirable features required for an expression vector are
segregation stability and the stability in the host for many generations under low selective
pressure (Keasling et al., 1999). Through this study, three classes of plasmids have been
employed based on the YEp, YCp and YIp vector series according to their destination of use.
5
Bioprospecting: “Searching and borrowing useful genes from other organisms to confer a specifically desired
phenotype” (Alper et al., 2009)
14
The YEp vectors, based on the 2 sequence are maintained at high copy number (< 7) in the
cell (Chen et al., 2012) and were applied to achieve high level of expression of the gene
encoding the enzyme catalyzing the final reaction toward the target product to ensure that this
step would not limit the entire process. (see Chapter 3.1.1). Differently, YCp vectors, based on
the CEN/ARS autonomous replication and centromeric sequence are maintained at low copy
number (1-2) in the cell (Fang et al., 2011). Due to the great level of segregation stability
provided and low metabolic burden they were employed for the reconstruction of synthetic
pathways (see Chapter 3.5). YIp integrative vectors on the other hand, do not replicate
autonomously and represent a versatile tool for rapid chromosomal integration; here they were
used to perform the promoter replacement studies (See Chapter 3.2.3).
Alternatively, classic PCR fragment-based genomic integration was applied in chromosome
engineering for gene deletion and stable gene expression during pathway optimization. For
gene overexpression applications, chromosomal integration offer the most stable solution. The
integration locus may however affect the expression level. In this study, previously characterized
integration sites were used in order to ensure the desired level of expression (Flagfeldt et al.,
2009). Multiple rounds of targeted sequential integration strategies based on recyclable
selectable markers for selection were employed for deletion/overexpression procedures (see
Chapter 3.3).
In an ideal context the platform strain would provide high genetic stability and ensure the
flexibility to allow the production of a range of different sequiterpene derivative compounds. In
order to combine these features in this study the functions required to redirect the carbon flux
toward the target pathway were integrated into the yeast genome, whereas the steps for the
final conversions were expressed on plasmids using the techniques described above.
2.4.2. Engineering transcription
Promoters represent a key determinant to transcriptionally control gene expression. Promoter
replacement techniques are an effective tool to control the gene expression at the
transcriptional level. Mainly two classes of promoter are utilized for pathway engineering,
constitutive and regulatable (inducible/repressible) expression systems. Strong constitutive
promoters have been widely applied to reach high levels of expression of target genes.
However, in some cases only small changes of expression are required; therefore the selection
of proper promoter systems is a critical choice to achieve the desired expression level in the
cultured cell. In order to achieve optimal transcription, several systems-orientated approaches
have been used to create synthetic promoter libraries of constitutive promoter with a wide range
15
of strength (Blount et al., 2012; Braatsh et al., 2008; Nevoigt et al., 2006; Alper et al., 2005,
Solem et al., 2002). Regulatable promoters instead are required when it is necessary to time the
gene expression during a determined process phase. Ideally a linear and uniform response to
the inducer/repressor concentration is preferable to achieve tight regulation. Some inducible
system in fact are affected by non uniform cell response that produces population heterogeneity
and may subsequently lead to a detrimental effect on cell growth affecting the overall
productivity (Keasling et al., 2007; Keasling et al., 1999).
In many cases transcript levels display context dependency. Different growth conditions,
medium and carbon source lead to different expression levels. For this reason, many studies
focus on characterizing and standardizing panels of promoters under multiple environmental
conditions to fine tune gene expression for pathway engineering applications (Sun et al., 2012;
Lee et al., 2011b; Kelly et al., 2009). In this study, both constitutive and regulatable promoters
have been applied and a simple screening method to titrate the promoter activities under the
desired condition has been developed.
Alternatively to promoter engineering, transcription factors - due to their global regulation role have been targeted in many studies for transcription level engineering using rational (Nielsen,
2001; Blom et al., 2000) and global approaches (Auslander et al., 2012; Alper et al., 2006). In
this thesis a modified version of a transcription factor known to regulate the targeted MVA
pathway was over-expressed to override the native host regulatory system.
2.4.3 Engineering translation-RNA processing
Driven by the development of inexpensive and rapid DNA synthesis procedures, de novo gene
synthesis for pathway engineering has become an economically feasible routine in many
laboratories. The novel synthesized genes are transferred into specific host strain to confer new
functionality; the expression of exogenous functions can be optimized at the translational level.
Recently, a great number of post transcriptional tools based on RNA control systems have been
developed e.g. asRNAs, miRNAs, siRNAs, rasiRNAs, riboregulators and riboswitches (Bayer et
al., 2005; Zamore et al., 2005; Isaacs et al., 2004; Patel et al., 1997). In this study codon
optimization methods and the use of antisense RNAs (asRNAs) have been applied. Codon
optimization successfully succeed in improving the rate of translation in many cases of foreign
gene expression in a heterologous host and appears to be particularly important when the
expressed function are sheared between microorganisms distantly related (e.g. as in the case of
this study C. lansium plant genes expressed in yeast S. cerevisiae). Several algorithms exist to
formulate codon optimization, however, unique design principles are yet not available. In the
future, application of synthetic biology to such guiding principles may play an important role in
16
generation of guidelines to overcome this crucial problem (Welch et al., 2009). Antisense RNAs
are a class of RNA regulatory molecules that control gene expression post-transciptionally
(Good, 2003). Antisense-based strategies consist of the use of an antisense RNA to bind a
target RNA sequence and e.g. inhibit translation. The expression of antisense copies of genes
has been used especially for plants genetic manipulations as an alternative to gene knockout
(Bourque, 1995), but only few applications of this technique are reported in the yeast S.
cerevisiae (Bonoli et al 2006; Olsson et al., 1997). In this study, an RNA-mediated strategy was
employed using a selected antisense DNA fragment comprising the 5’ region of the target gene
and part of its 5’UTR, controlled under a specific promoter to express an mRNA antisense
construct for silencing the target gene (see Chapter 3.2.3).
2.4.4 Engineering post translation
Protein engineering for pathway engineering is a vast area of research that recently gained
benefit from the application of computational techniques (Keith et al., 2007). A large number of
protein-level regulatory mechanisms exist to control protein function, activity, stability and
localization. Much of the effort in protein manipulation methods focuses on modifying protein
properties
(e.g.
Vmax,
K m,
cofactor/substrate/product
specificity)
to
improve
catalytic
performances (Leonard et al., 2010, Watanabe et al., 2007; Yoshikuni et al., 2006). In contrast,
to target catalytic proprieties, simple examples of protein level engineering are based on
modifying protein regulatory functions and their localization (Steen et al., 2010; Cho et al.,
1995). In this study, a key regulatory enzyme of the targeted pathway was re-localized
expressing a truncated form of the protein deleted in the periplasmic membrane anchor domain
resulting in a cytolsolic soluble form that bypasses the endogenous regulatory feedback loops
(see Chapter 3.2.2)
Beyond these reported approaches a number of elegant protein-based solutions for pathway
engineering have been recently demonstrated e.g. direct protein fusion strategies (Albertsen et
al., 2010), synthetic scaffold systems (Dueber et al., 2009), protein shell systems (Lee et al.,
2011c) and tag localization in cellular sub-compartment (Farhi et al., 2011), focused to localize
engineered functions and spatially organize pathways. Although these technique represent an
active growing branch of pathway engineering and they have been successfully applied in
several cases, they are not the primary focus of this thesis and will therefore not be further
discussed.
17
2.5 Production process design-Industrial microbial fermentation
Microbial high density fermentation capabilities make industrial-scale sesquiterpene production
attractive in a prospective of a viable biotechnological production process. The development of
an efficient bioreactor operation has great impact in the optimization of a competitive
bioprocess (particularly in the case of low-value products), process engineering plays a critical
role in the establishment of a low-cost process (Leib et al., 2001). Essentially three different
reactor configurations are applied in industrial production processes: (i) batch, (ii) fed-batch
(including its variant repeated fed-batch) and (iii) continuous (Nielsen et al., 2003) (Fig 2.3). The
different operations modes are briefly discussed below referring specifically to yeast S.
cerevisiae cultivation cases; only the stirred tank reactor, which is the workhorse of the
fermentation industry, is considered.
Figure 2.3. Simplified scheme of the three principal cultivation modes employed during
a biotechnological process. Batch (FIN= FOUT= 0); Fed-batch (FIN≠ 0; F = 0) and
OUT
continuous (FIN=FOUT≠ 0) process details are described in the text. The different phases
which the cell undergoes during the process are highlighted. Adapted from (Nielsen et al.,
2003; Stephanopoulos et al., 1998.; Weusthuis et al., 1994; Heijnen et al., 1992).
2.5.1 Batch Cultivation
The batch method is the simplest cultivation technique, pH and dissolved oxygen (DO) are
controlled, carbon source (generally sugar) and the required nutrients are provided in excess at
the beginning of the cultivation and the fermentor working volume is constant during the entire
process (FIN= FOUT= 0). Typical exponential growth is achieved that proceeds at the maximum
rate attainable (µmax). When glucose is used as substrate in aerobic conditions yeast metabolism
is respiro/fermentative where glucose is mainly fermented to ethanol. After complete sugar
18
consumption, the diauxic shift occurs and the fermentation byproducts accumulated in the first
phase (ethanol, acetate and glycerol) are re-consumed. The diauxic growth is the result of
carbon catabolite repression. Due to the easy set-up batch culture is an essential tool for
preliminary screening of strain physiology.
2.5.2 Fed-Batch Cultivation
The majority of industrial processes are nowadays carried out using fed-batch cultivation
methods. The process initiates as batch and after a suitable amount of biomass is obtained a
feed of fresh concentrated medium is applied but no volume is withdrawn from the fermentor
resulting in an increase of the working volume with time (FIN≠ 0; FOUT= 0). The feed strategy
applied influences the overall process performances. Typical glucose based feed configurations
are based on a first phase were the feed is kept exponential and a second phase when high cell
concentration is reached with constant feed rate to avoid potential limitations (Pham et al.,
1998). Ideally the process proceeds maintaining the sugar concentration below the critical level
preventing the Crabtree effect, maintaining a respiratory metabolism and avoiding the switch to
fermentative metabolism. Advances in fermentation technology produced a multitude of
strategies focused on proper control of the feed addition in order to avoid the detrimental
effects due to over/under feeding (Lee et al., 1999). An improved variant of the fed-batch
consist in a repeated fed batch system were at the end of the fed-batch process a certain
volume of culture is periodically withdrawn from the system (FIN≠ FOUT≠ 0) (Heijnen et al., 1992).
The main advantage of using fed-batch in a large scale process is the high final titer achievable.
During this study, an optimized fed-batch production process was designed for sesquiterpene
bio-production. Additionally, a feed control method for optimizing the production process was
developed (see Chapter 3.4.2).
2.5.3 Continuous Cultivation
In continuous cultivation mode, also commonly called chemostat, the process starts as a batch
similarly to the fed-batch set up. Thereafter follows constant addition of fresh media at a fixed
rate and continual removal of spent medium at the same rate, maintaining the working volume
constant (FIN=FOUT≠ 0). After some time the cells will reach a “steady state” growth condition.
Cell growth is usually controlled using a single limiting nutrient (generally the carbon source). In a
glucose limited chemostat yeast metabolism is fully respiratory and sugar is completely oxidized
to biomass and carbon dioxide as the major products, while fermentation products are absent.
Under ideal conditions the growth rate is equal to the dilution rate (D) imposed, and the
chemostat cultivation therefore allows to change the operational specific growth rate
19
(independently of the other parameters) by varying the feed flow to the reactor. The maximum D
applicable (Dcrit) corresponds to the µmax (obtained in batch) and for higher dilution rates a wash
out occurs (D> Dcrit). Typical industrial yeast continuous culture applications are carried out at
D= 0.1 h-1 or greater to allow a productivity advantage versus batch culture (Heijnen et al.,
1992). Chemostat cultivation methods have been applied in this study as a tool to investigate
the sesquiterpene productivity of the genetically engineered strains constructed, and a novel
chemostat set-up production method that allowed for continuous product recovery and suitable
for industrial scale up was developed (see Chapter 3.4.5).
2.6. Techno-economical analysis of sesquiterpene microbial production
Development of a cost competitive bio-production requires a detailed analysis of the production
process performances. The titer6, yields7 and productivities8 of the target compound are an
important set of parameters to monitor for optimization of the fermentation process (Nielsen et
al., 2002). During the development of a microbial production process different aspects including
physicochemical proprieties of the target compounds and the formation pathway have to be
carefully analyzed. Because the final costs of the process depend in large amount on the
conversion of the substrate, one of the first parameters to take into consideration is the maximal
theoretical yield Ysp. This value cannot be overcome and corresponds to the highest possible
product amount achievable from a certain amount of substrate and it can be expressed as
Cmol product Cmol substrate-1. Ysp for α-santalene from different carbon sources can be
calculated as follow: Ysp= κs/κp from a simple energy balance assuming that all the energy
content of the substrate (electrons) ended up in the product, where the degree of reduction
(DOR) of substrate (κs) and the product (κp) gives Ysp. The reduction level express in 1 C-atom
bases and Ysp of the target compound
α-santalene from different substrates is reported in table
2.2.
6
Titer: Final measure of the product concentration
7
Yield: Efficiency of substrate conversion to product
8
Productivity: Volumetric production rate, mass of compound produced per unit weight of cell per unit time
20
Table 2.2. α-santalene maximal theoretical yield and pathway yield under different carbon
sources
Formula
Formula
(1
C‐atom)
Degree
of
reduction
per
carbon
κ Ysp
‐
(Cmol
Cmol
1
)
Y
‐1
(Cmol
Cmol )
Glucose
C6H12O6
CH2O
4
0.86
0.56
Xylose
C5H10O5
CH2O
4
0.86
0.56
Ethanol
C2H6O
CH3O1/2
6
1.29
0.83
α‐Santalene
C15H24
CH5/8
4.625
‐
‐
Compound
Substrate
Product
p
The calculation of Ysp is based only on the substrate/product analysis and it is independent of
the metabolic network. However, in the early process stage it is useful to determine the
economic feasibility of the process simply based on the substrate cost and product income
determining the maximum usable energy contained in the substrate that can be transferred to
the product.
Analysis of the metabolic pathway allows determining the stoichiometric equation for product
formation and its redox balance to evaluate the efficiency of the product synthesis through a
specific pathway. In the case of α-santalene production in S. cerevisiae from different substrates
(glucose eq. 1; xylose eq. 2; ethanol eq. 3) via the MVA pathway at purely oxidative growth9 it
can be summarized as follow:
- CH2O - 1/3 ATP - 2/9 NADPH + 5/9 CH8/5 + 2/3 NADH + 4/9 CO2 = 0
(1.1)
- CH2O - 1/5 ATP - 19/45 NADPH + 5/9 CH8/5 + 13/15 NADH + 4/9 CO2 = 0
(1.2)
- CH3O1/2 - 1 ATP - 1/3 NADPH + 5/6 CH8/5 + 1/2 NADH + 1/6 CO2 = 0
(1.3)
Pathway analysis results in a α-santalene product yield of Yp= 0.56 Cmol Cmol-1 for glucose and
xylose and Yp= 0.83 Cmol Cmol-1 for ethanol, respectively, corresponding to a reduction of 35%
(glucose & xylose) and 36% (ethanol) compared to the maximum yield achievable (Table 2.2).
In all the three cases NADPH and ATP is required for product formation and an excess of NADH
is produced. If it is assumed that neither ATP nor cofactors NADH and NADH can accumulate
in the cell, an energy balance can be calculate accounting for the required amount of substrate
to compensate the pathway’s redox imbalance.
9
Calculations are made assuming that during oxydative conditions the formation of cytosolic acetate produced in the
reaction catalyzed by acetaldehyde dehydrogenase (ACDH) uses NAD as exclusive cofactor leading to the formation
of 1 molecule of NADH per molecule of acetate produced (Frick et al., 2005).
+
21
CHAPTER 3 Results & Discussion
3.1 Construction of a yeast “sesquiterpene cell factory”: α-santalene case study
The main objective of this research was the construction of an efficient S. cerevisiae cell factory
capable to produce industrially relevant titers of the sequiterpene hydrocarbon α-santalene, a
precursor for commercially interesting compounds.
3.1.1 Minimal engineering of yeast for sesquiterpene production: expression of a heterologous
plant gene in S. cerevisiae
The first limit in the construction of a yeast cell factory for sesquiterpene production relies on the
ability to efficiently express a heterologous plant sesquiterpene synthase. The target compound
of this study, α-santalene, is produced in a one step reaction from FPP enzymatically catalyzed
by plant santalene synthase. α-Santalene structurally related sequiterpene compounds are
widely present and conserved in plant species, and analysis of Clausena lansium (wampee)
leaves identified a high content of α-santalol (Zhao et al., 2004; Pino et al., 2006). The santalene
synthase gene (SanSyn) employed in this study was identified through a cDNA library screening
from C. lansium and was specifically selected due to its previously demonstrated high specificity
of 92% towards production of α-santalene (Schalk, 2011). Santalene synthase (SNS) belong to
the class I group of sesquiterpene cyclases that are among the most studied terpene synthase
(Christianson et al., 2008). These enzymes catalyze a complex intermolecular cyclization of FPP
with very different product specificity and the reaction mechanism often involves several partial
reactions (Fig 3.1). Conversions of the linear FPP into cyclic derivatives are not trivial as it may
appear and involve limited numbers of mechanisms dictated from the FPP trans-geometry of
the double bond and result in the production of diverse classes of sesquiterpenes; FPP
cyclization to α-santalene occurs via an enzyme bound nerolidyl diphosphate intermediate
(NPP). The substrate is bound in the enzyme’s hydrophobic pocket that determines the
stereochemistry of the product. The reaction is initiated by the carbocation formation via loss of
the diphosphate group (OPP-), which is kept in complex with Mg2+, and subsequent
rearrangements define the final product and determine the specificity of the enzyme. Fast OPPrelease can stop the reaction and result in alternative products (Fig. 3.1) (Jones et al., 2011;
Christianson et al., 2008; McCaskill et al., 1997).
22
Figure 3.1. Detailed reaction mechanism of plant santalene synthase (SNS). Electrophilic attack on the central
double bound of the substrate (E,E)-farnesyl diphospahte produces an allylic carbocation that can evolve into
formation of linear product (E,E)-farnesene or one of the cyclic derivatives α-santalene and trans-αbergamotene via a nerolidyl diphosphate intermediate (NPP). Adapted from McCaskill et al., 1997; Christianson et
al., 2008 and Jones et al., 2011.
In order to ensure high santalene synthase levels an expression vector with suitable
transcriptional promoter/terminator was chosen and constructed (Partow et al., 2010).
Introducing SanSyn yeast was minimally engineered for the first time to produce α-santalene.
Product analysis revealed that α-santalene was the main product detected with 1.45 mg l-1 and
only a minor amount, 0.17 mg l-1, of the secondary product trans-α-bergamotene was found.
During bio-production the product purity and quality is a major driver to meet commercial
demands. The structure of the sesquiterpene produced estimated by GC/MS was identical
(∼98% purity) to the one produced in plant (Fig. 3.2).
Many studies have reported successful examples of heterologous production of isoprenoids by
simply expressing plant synthase genes in a desired microbial host. Not surprising the resulting
yield of this simple straightforward approach was often extremely low (ranging between 0.038
and 6.7 mg l-1) (Farhi et al., 2011; Wang et al., 2011b; Asadollahi et al., 2008; Paradise et al.,
2008; Ro et al., 2006; DeJong et al., 2005; Jackson et al., 2003; Madsen et al., 2001).
23
Figure 3.2. (A) Total ion chromatograms from GC-MS analysis of authentic standard of
farnesol, α-santalene, and an extract of engineered S. cerevisiae showing peaks of αsantalene (S), trans-α-bergamotene (B) and E,E-farnesol (F). The representative ion
chromatogram referred to as yeast products was obtained during ISPR fed batch
fermentation (for cultivation methods details see Chapter 2.5). (B) Mass spectra and
retention times of α-santalene produced from yeast and extracted from plant (left panel)
and E,E-farnesol produced from yeast and chemical standard (right panel).
24
The catalytic efficiency (Vmax/Km) and the specificity are often referred to as key factors during
heterologous production (Picaud et al., 2005). Subsequently, during this study a codonoptimized artificial santalene synthase (SanSynOpt) for optimal expression in S cerevisiae was
designed. Expression of the codon-optimized SanSynopt led to comparable specificity and only
modest increase in efficiency compared with the wild type version, suggesting that although the
codon bias has an important role, the level of expression depends on multiple proprieties and
other factors may be critical (e.g. mRNA stability, sequence that control the initiation of the
translation, nucleotide sequence surrounding the N-terminal region, tRNA levels) (Gustafsson et
al., 2004).
3.2 Rationally designed metabolic control engineering approach
A second bottleneck that often limits the production of a heterologous compound is the
capacity to increase the precursor pool in order to enable efficient conversion to the target
compound. Yeast has a very limited secondary metabolism and terpenes are produced
exclusively through the mevalonate pathway (see Chapter 1.2). Due to the variety of essential
compounds produced in the MVA pathway, the activity of many enzymes of this pathway is
strictly regulated at different levels (Maury et al., 2005). A rationally designed metabolic control
engineering approach was employed to maximize flux through the MVA pathway and obtain
optimal sesquiterpene production. This approach relies on the deep knowledge available of
yeast biology and MVA pathway regulation. Two of the well recognized regulatory steps of the
MVA pathway catalyzed by 3-hydroxy-3-metyl-glutaryl-coenzyme A reductase (HMGR) and
squalene synthase (SQS) were optimized by introducing genetic modifications that enable to
channel increased flux towards α-santalene synthesis.
3.2.1 Engineering the regulatory checkpoint of the MVA pathway.
α-Santalene production was increased combining (i) de-regulating the MVA pathway
overexpressing a truncated version of HMG-CoA reductase (tHMG1) and (ii) dynamic control of
the MVA pathway branch point by down regulating the squalene synthase gene (ERG9) (Fig.
3.3).
25
Figure 3.3. Metabolic engineering strategy for overproducing α-santalene. Two key checkpoints in the MVA
pathway were engineered. (i) The rate controlling step catalyzed by HMGR was de-regulated to maximize the flux
through the MVA pathway overexpressing a truncated non-membrane bound version of HMG1 that represents a
constitutively active form of HMGR. (ii) Enzymatic activities acting at the FPP branch point were modulated to redirect
carbon flux towards the desired target compound; the main FPP consuming reaction SQS was down-regulated using
a promoter replacement technique and two activities competing for FPP, Lpp1 and Dpp1, were disrupted.
26
3.2.2 De-regulation of the MVA pathway to increase the critical precursor pool
As previously mentioned, because of its crucial roles in supply of several essential compounds
the MVA pathway has evolved a hierarchical control architecture. De-regulation is therefore
necessary to increase flux trough this pathway to increase the precursor pool for isoprenoid
synthesis. The conversion of 3-hydroxy-3-metyl-glutaryl-CoA into mevalonate catalyzed by
HMGR is probably the most studied enzyme of all and it is considered to exert a high degree of
MVA flux control (Scallen et al., 1983; Basson et al., 1986). In yeast two isoform of HMGR exist
and their activity is subject to extensive regulation including feedback regulation and crossregulation (Hampton et al., 1996, 1994; Brown et al., 1980). HMGR is composed of an
interspecies conserved catalytic domain and a variable membrane anchoring N-terminal region
referred to as sterol sensing domain (SSD) that spans the membrane of the endoplasmic
reticulum (ER) and interact with sterol sensing components of the ER membrane. Part of Hmg1
regulation acts through a complex mechanism leading to protein degradation at the level of the
N-terminal domains (SSD domain) (Nielsen, 2009). Overexpression of the truncated form
containing only the catalytic domain and lacking the regulatory domain bypasses this post
transcriptional circuit and results in a constitutively active soluble form that is non-membrane
bound (Fig. 3.3) (Polakowski et al., 1998; Donald et al., 1997). The use of the deregulated form
of Hmg1 (tHmg1) represents an excellent example of bypassing the regulatory mechanisms
controlling the MVA flux and has been successfully applied to a series of microbial production
processes to increase the flow through the pathway (Fahri et al., 2011; Asadollahi et al., 2010,
2009; Kirby et al., 2008; Ro et al., 2006; Jackson et al., 2003).
Previous studies demonstrated that a high level of expression is required to ensure a high MVA
flux (Ro et al., 2006), and this strategy was therefore applied by constructing a high copy
number expression vector containing tHMG1 and SanSyn under control of strong promoters
and this resulted in a 2 fold increase in sesquiterpene production yielding 3.1 mg l-1 α-santalene
and 0.33 mg l-1 trans-α-bergamotene.
3.2.3 Dynamic control of MVA pathway branch point
The second MVA flux controlling step is represented by SQS that regulates the FPP flux
distribution between sterols, e.g. lanosterol, erosterol, and non-sterols, e.g. dolichols,
ubiquinone, heme A, prenyated proteins, and sesquiterpene derived products. FPP is a pivotal
intermediate and its intracellular concentration is carefully regulated by a flow diversion
mechanism. Under normal growth conditions the cellular sterol demand is higher than that of
non-sterols, and most of the FPP is converted into ergosterol and SQS is therefore the main
27
FPP consuming reaction (Kennedy et al., 1999). In order to minimize the overflow to
biosynthetically related sterols optimization of FPP branch point flux distribution is necessary.
Deletion of the ERG9 gene encoding SQS produces lethal mutants because of the essential role
of ergosterol in maintaining the membrane fluidity (Jennings et al., 1991) and restoration of an
erg9Δ mutantion would require ergosterol supplementation that would have consequences on
the economic feasibility of the entire process (Takahashi et al., 2007). Therefore a suitable
approach to increase the FPP availability for conversion into α-santalene is to reduce the flux
through SQS enabling sufficient squalene to satisfy the minimum amount of ergosterol
necessary to fulfill cellular growth. Precise adjustment of an essential enzymatic activity avoiding
unbalance represented a challenging task to overcome. A variety of tools has been developed
to modulate yeast gene expression (see Chapter 2.4). Among the several genetics techniques
available as alternative to complete gene deletion in order to reduce specific gene activity
(Hammer et al., 2006; Mjiakovic et al., 2005) promoter replacement and the use of
repressible/inducible promoter systems (Kaufmann et al., 2011) represents an efficient strategy
to transcriptionally fine tune gene expression. Previous attempts to regulate SQS activity were
mainly based on replacement of the native ERG9 promoter (PERG9) with a methionine-repressible
promoter system (PMET3) (Asadollahi et al., 2008; Paradise et al., 2008; Ro et al., 2006).
Promoter
PHXT1
Description
glucose
concentration
controlled
promoter
of
the
hexose
transporter
gene
HXT1
Reference
Ozcan
et
al.,
1995
Lewi
et
al.,
1991
PHXT2
the
HXT2
promoter
for
gene
silencing
approach
expressing
ERG9
antisense
mRNA
Ozcan
et
al.,
1995
PTEF1M2
Low‐level
constitutive
TEF1
promoter
mutant
was
selected
after
directed
evolution
approach
based
on
error
prone
PCR
Alper
et
al.,
2005
Nevoigt
et
al.,
2006
Figure 3.4. Characterization of candidate promoter
strength during shake flask cultivation in fed-batch
mode. ß-Galactosidase activity with the different
tested promoters is the average of values obtained
from at least three independent cultivations assayed
in duplicates.
Table 3.1. Candidate promoter systems and their
brief function description evaluated to promote the
activity of SQS.
Ideally the level of repression should be proportional to the concentration of the inducer
provided; indeed a careful evaluation of PMET3 activity conducted during this study using lacZ
28
reporter system revealed severe difficulties in controlling the promoter activity mainly related to
the ability of the cell to metabolize the repressing agent methionine. Moreover this system
suffers from several limitations for the industrial scale development due to the cost of the
repressor itself.
Using a metabolic control engineering approach a dynamic controller able to sense the
physiological state of the cell (fermentative/ fully respiratory metabolism) and regulate the
expression of SQS was constructed using a promoter replacement technique. A series of
alternative promoter systems was preliminary screened for their activity using the reporter LacZ
gene assay in order to identify a suitable promoter that could provide optimal expression of
SQS.
The ERG9 gene expression level is
context
dependent
and
changes
based on the growth condition and
carbon source utilized (Kennedy et al.,
2001, 1999). An efficient screening
method was developed capable to
simulating the growth condition typical
of an industrial production process.
The promoters were evaluated in a
glucose limited process where the
carbon source (glucose) is released
using
silicone
controlled
elastiomers
kinetics.
This
with
screening
method mimics a typical fed-batch
process
widely
applied
industrial
production
(see
during
Chapter
2.5.2) and allowed precise comparison
Figure 3.5. (A) Ergosterol production rate (mg g biomass-1 h-1). (B) αsantalene and E,E-farnesol production rate in S. cerevisiae strains
engineered for sesquiterpene production. P E R G 9 native ERG9 promoter
serve as reference strain; in P H X T 1 the ERG9 promoter had been replaced
with HXT1 promoter; P H X T 2 carried an antisense DNA fragment
comprising the 5’region of ERG9 and part of its 5’ UTR, whose
expression is controlled by HXT2 promoter; P M E T 3 carried the methionine
repressible promoter instead of the native PERG9, for this strain Lmethionine was added at regular intervals in order to maintain ERG9
repressed. Strains were growth in two phase partitioned fed-batch
glucose limited cultivation mode. The error bars represent the standard
deviation for two independent cultivations.
of the candidate promoters under the
selected condition.
Two regulatory systems PHXT1 and PHXT2
were chosen based on the promoter
characterization study according to
their expression profile. Sequiterpene
producing
strains
containing
the
expression vector (expressing tHMG1 and SanSyn) were engineered at the SQS level using
29
PHXT1 and PHXT2 to promote the expression of ERG9 and antisense ERG9, respectively, during a
low glucose level typical of a fed-batch cultivation mode with the aim to couple SQS activity to
glucose concentration and achieve maximal repression during the feed phase when glucose is
limiting. The selected replacement mutants together with wild type and previously widely applied
PMET3 system were evaluated for their sesquiterpene production capacity and the total sterol
content (Fig 3.5). Applying different levels of repression of SQS result in a redirection of carbon
flux from ergosterol towards sequiterpene products. From the obtained data it was clear that
engineering the FPP branch point modulating SQS activity resulted in an increase in FPP that
was redistributed toward sesquiterpene products. The best α-santalene producer expression
systems selected was PHXT1 that is regulated by limiting nutriment (extracellular glucose
concentration) and allows timing the promoter strength changing the cultivation condition,
separating growth phase (excess of extracellular glucose -> fermentative metabolism) and
production phase (limited glucose -> respiratory metabolism) maintaining the repression at the
set level during the desired phase without addition of repressing agent. Moreover, using this
technique the cellular response is uniformly distributed in the population avoiding the creation of
suboptimal population with different expression levels.
The sesquiterpene pool produced from the engineered mutant tested was composed of αsantalene as major product, trans-α-bergamotene and E,E-farnesol (FOH). The amount of transα-bergamotene was produced proportionally to α-santalene and corresponds to 12% of the
total α-santalene confirming that this compound is a secondary product of SNS (as previously
anticipated). Substantial accumulation of FOH, a FPP-derived product was observed in all the
engineered strains. FOH is the result of the dephosphorylation of FPP. In yeast, specific
enzymatic activity involved in this process has not yet been elucidated. SQS has a lower affinity
for FPP (Km=2.5 M Vmax=0.46 pmol min-1 mg-1) compared to most of the other enzymes acting
at the FPP branch point (Km= 0.42-0.55 M Vmax=0.86-2.1 pmol min-1 mg-1). This produces a
high flux toward non-sterol products when the intracellular FPP concentration is low (Scheffler et
al., 2002). Obtained results suggest that SNS was capable to compete with the non-sterol
branches and drain FPP toward sesquiterpene compounds. However, when the level of FPP
was increased by down-regulating SQS the catalytic performances of SNS were not able to
completely convert the excess of FPP to α-santalene and there was therefore overflow with
dephosphorylation of FPP resulting in formation of FOH. The FOH overflow reduces substrate
availability for SNS, and in order to reduce the competition for the same substrate a
complementary strategy was investigated in this study. Two identified phosphatases
responsible for most of the cytosolic isoprenoid phosphatase activity, Lpp1 and Dpp1, were
deleted. The single (lpp1Δ) and double deleted mutant (lpp1Δ dpp1Δ) further improved α
30
santalene production at the expense of FOH formation. Reduction of FOH formation confirms
the role of these two lipid phosphatase in FPP dephosphorylation, the incomplete reduction of
FOH formation, however, suggests that other cellular mechanisms are involved in this
conversion. This first round of strain engineering resulted in the production of a base strain
capable of a final α-santalene productivity and titer equal to 0.015 Cmmol gDCW-1 h-1 (= 0.21
mg gDCW-1 h-1 ) corresponding to 92 mg l-1.
3.3 Combined metabolic engineering strategy of precursor and co-factor supply for
sesquiterpene production.
One of the major challenges in order to develop an efficient bioprocess is identifying the rate
limiting regulatory steps and designing a system wide process avoiding imbalances of pathway
intermediates and co-factors. A second round of strain optimization was performed where a
multistep metabolic engineering strategy was designed combining four different approaches
(Box 3.1).
Box.
3.1.
Multistep
metabolic
engineering
strategy
for
the
construction
of
a
S.
cerevisiae
“sesquiterpene
cell
factory”
(i)
Modulation
and
optimization
of
the
FPP
branch
point
(PHXT1ERG9
&
ΔLPP1
ΔDPP1).
(ii)
De‐regulation
of
MVA
pathway
to
increase
the
precursor
pool
for
isoprenoid
synthesis
(↑tHGM1
&
↑ERG20).
(iii)
Increasing
the
availability
of
the
reductive
cofactor
NADPH
modifying
the
ammonium
assimilation
pathway
(ΔGDH1
&
↑GDH2).
(iv)
Enhancing
the
activity
of
a
transcriptional
activator
of
sterol
biosynthesis
(↑upc2‐1).
(i) The first approach was focused on
minimizing the overflow to biosynthetically
related
sterols
that
have
the
same
precursor as target product and was
explored in details during the 1st round of
strain engineering (Chapter 3.2.2). The
best producer was selected (PHXT1) and
utilized
attempts
availabilty
during
this
to
increase
resulted
stage.
Previous
cytosolic
in
a
FPP
rapid
dephosphorylation of FPP to FOH, and to
minimize FOH overflow Lpp1 and Dpp1
activities were deleted (Δlpp1 & Δdpp1) (ii)
A second part of the strategy consisted of amplifying the flux through the MVA pathway, and
this was realized combining the previously described over-expression of a constitutively active
cytosolic variant of Hmg1p (tHMG1) with the over-expression of a second enzyme in the MVA
pathway, farnesyl diphosphare synthase (FPPS), catalyzing the multiple condensation of IPP
units into FPP (Fig. 1.1). Due to the pivotal nature of the FPP molecule its synthesis and
distribution into derived products is strictly regulated by FPPS (Chambon et al., 1990;
Grabinska et al., 2002). The effect of ERG20 (gene encoding FPPS) over-expression was
therefore investigated in addition to over-expression of tHMG1. (iii) The manipulation of the
31
NADH and NADPH co-factor balance in order to overcome limits imposed from the cellular
redox constraints is a well-established metabolic engineering strategy (Hou et al., 2010). The
reaction leading to α-santalene formation results in net production of NADH and consumption of
NADPH (see Chapter 2.6). A change in the NADH:NADPH ratio in favor of NADPH would
therefore be beneficial for product formation. In silico analysis identified the deletion of the
NADPH consuming reaction of glutamate dehydrogenase (encoded by GDH1) as a target
strategy to increasing the availability of the reduced co-factor NADPH (Asadollahi et al., 2010).
Activation of an alternative ammonium utilization route in a GDH1 deleted strain by overexpressing the NAD-dependent glutamate dehydrogenase (encoded by GDH2) resulted in an
increase of NADH consumption during the anabolic process and in a modification of the yeast
co-factor balance (dos Santos et al., 2003). Here, the effect of GDH1 deletion alone as well as
coupled with simultaneous over-expression of GDH2 on α-santalene production was evaluated.
(iv) The last strategy employed involved engineering of a key transcription factor with the
objective to generally up-regulate expression of the MVA pathway genes. Upc2 and Ecm22
have been identified as the main transcription factors responsible for the activation of several
MVA and ergosterol pathways genes (Vik et al., 2001). The point mutation upc2-1 discovered
first for conferring the ability to assimilate extracellular sterols during aerobic cultivation (Lewis et
al., 1988) has been demonstrated to result in a constitutively active form of Upc2 (Davies et al.,
2005). Over-expression of upc2-1 has been employed to transcriptionally up-regulate the MVA
pathway genes during isoprenoid production.
32
Figure 3.6. Overview of the multistep genetic engineering approach for increasing α-santalene production. (A)
Expression plasmid containing tHMG1 encoding truncated HMG-CoA reductase, a codon optimized santalene
synthase gene (SanSyn ) P and P promoters as well as T and T terminator sequences. (B) Integrated
cassettes, rectangles containing arrows represent the promoters and their directionality, pentagons the genes and
empty squares the terminators. (C) Scheme of the engineered mevalonate, prenyl phosphate and ammonium
assimilation pathways and FPP branch point; overexpressed and deleted genes are highlighted. Overexpressed
genes are tHMG1 (encoding truncated HMG-CoA reductase), ERG20 (encoding FPP synthase), GDH2 (encoding
NAD-dependent glutamate dehydrogenase), and SanSyn (encoding α-santalene synthase). Deleted genes are
GDH1 (encoding NADP-dependent glutamate dehydrogenase), LPP1 and DPP1 (both encoding lipid phosphate
phosphatases). The promoter of the ERG9 gene (encoding squalene synthase) is replaced with P . Genes whose
promoters contain Upc2 binding sites are indicated with a grey arrow: ERG13 (encoding HMG-CoA synthase),
ERG12 (encoding mevalonate kinase), and ERG8 (encoding phosphomevalonate kinase). Additional genes
indicated are ERG10 (encoding acetoacetyl-CoA thiolase), ERG19 (encoding diphosphomevalonate decarboxylase)
and IDI (encoding IPP isomerase).
opt
TEF1
PGK1
ADH1
CYC1
opt
HXT1
33
All genetic modifications described
were
chromosomally
integrated
to
enhance genetic stability during long
term cultivations and avoid constant
application of a selective pressure for
maintaining the construct. In order to
obtain high level expression the genes
were targeted via sequential integration
strategy into suitable expression loci
previously characterized to confer high
gene expression (Flagfeldt et al., 2009).
Platform flexibility was maintained by
expressing the synthase gene for the
final
conversion
of
FPP
into
α-
santalene with the previously described
Figure 3.7. Sesquiterpene productivity in a two-phase
partitioned glucose-limited aerobic chemostat. αSantalene and farnesol production rate in Cmmol (g
biomass)-1 h-1 (the C-molar weight of α-santalene and
farnesol are, respectively, 13.62 and 14.82 g Cmol-1)
cultivated at dilution rate D=0.05 h-1. Error bars represent
the standard deviation from three independent cultivations
(top). List of S. cerevisiae strains engineered (bottom).
multi-copy
expression
allows
utilize
platform
to
for
a
the
range
vector.
This
constructed
of
different
isoprenoids. The contribution of the
different
strategies
and
their
combination on isoprenoid production
was evaluated. The impact of different combinations of the designed metabolic engineering
strategies on sesquiterpene productivity is summarized in figure 3.7. Replacing the native ERG9
promoter (PERG9-> PHXT1) and deleting LPP1 (strain SCIGS29) increased the yield and productivity
of α-santalene 3- and 3.8- fold, respectively, over the control strain (strain SCIGS28). Additional
DPP1 deletion (strain SCIGS30) reduced the rate of hydrolysis of FPP into the by-product FOH
2-fold leaving an unchanged flux towards α-santalene. Subsequent GDH1 deletion (strain
SCIGS31) led to no further enhancement in α-santalene productivity, but interestingly no FOH
was formed in this strain. When combined with GDH2 and ERG20 over-expression (strain
SCIGS24) a 4- and 6-fold improvement in α-santalene yield and productivity, respectively, was
respectively. Finally, including the overexpression of upc2-1 and having an extra copy of tHMG1
(strain SCIGS25) resulted in an insignificant change in α-santalene formation but a 2-fold
increase in FOH production. The optimal solution was obtained through combining all the
modification (except upc2-1) resulting in the highest α-santalene yield of 0.0052 Cmmol Cmmol1
and productivity of 0.036 Cmmol gDCW-1h-1 (Strains SCIGS24). The engineering strategy
34
allowed a 6-fold increase in α-santalene productivity compared to the control strain under the
tested conditions and highlights the importance of this systematic approach.
3.4 Development of an efficient fermentation process
During the realization of a microbial bioprocess, the overall aspect of the entire process should
be taken in consideration and strain construction should be performed in parallel with
optimization of fermentation and recovery procedures. The selected cultivation method and
consequently downstream steps are critical choices that affect the economic feasibility of the
final system. During this study, the strain improvement strategy was integrated with the
development of a cost effective fermentation process resulting in an efficient fermentation
strategy that couples biochemical production to biomass formation and improves α-santalene
production.
3.4.1 Fed batch in situ product removal (ISPR) integrated bio-process
Bioreactor operations have strong impact on design of an efficient bioprocess and influence in a
significant way the production performances. Fed batch fermentation is commonly used during
industrial production processes to achieve high titer, yield and productivity of the target
compound (Nielsen et al., 2003). Limited exponential feed profiles were used to maximize the
carbon flux from glucose to biomass and the desired compound, alleviating glucose repression
and Crabtree effect (Pronk et al., 1996). Here the feed rate was designed using an exponential
policy chosen so that the volumetric rate of biomass production was constant and equal to the
specific growth rate. Exponential feed rate v(t) (l h-1) was calculated according to equation 3.1.
ν (t)=
x0V0 exp (µ0t)
(3.1)
where x0, s0 and V0 were the biomass density (gDCW l-1), the substrate concentration (g l-1) and
the reactor volume (l) at the start of the feed phase, Yxs was the yield coefficient (g glucose
gDCW-1); sf was the concentration of the growth limiting substrate (g glucose l-1) in the reservoir;
µ0 the was the specific growth rate (h-1) during the feed phase and t the feeding time (Nielsen et
al., 2003).
Due to the low water solubility of the isoprene products this compound can easily be stripped
with the gas bubble used for aeration. An in situ product removal was therefore applied to
maximize the product recovery. This technique is extensively used in bioprocesses to produce
35
secondary metabolites (Stark et al.,
2003), and it consists of a double phase
partitioning system that allows constant
product removal from the cells and
trapping the product into the solvent
phase minimizing product loss through
the gas outlet and reducing potential
toxic
effects
accumulation
due
in
the
to
product
aquous phase.
Dodecane was selected as organic
phase
due
to
its
hydrophobicity
(logPdodecane: 6.6; LogPsantalene: 6.2), low
volatility and biocompatibility with S.
cerevisiae
(Asadollahi
et
al.,
2008;
Newman et al., 2006). The organic layer
was added before initiating the feed
addition corresponding to the start of the
production phase. This set-up allows
performing
extractive
fermentation
directly in the bioreactor resulting in an
integrated production/recovery process.
Analysis of process performance shows
Figure 3.8. Time course of an aerobic fed-batch culture
with exponential sugar feed of S. cerevisiae strains.
Cultivation was started as batch with 1 Cmol l-1 (30 g l-1) of
glucose as carbon source, after complete glucose
consumption and residual ethanol produced during the
glucose consumption phase was completely depleted
production phase was started feeding concentrated
substrate (7.4 Cmol l-1 glucose as carbon source) with
exponential kinetics for a feed period of 36 h. The feed of
glucose (ml h-1) is shown on the upper graph. Typical profile
observed for formation of biomass (g l-1, filled diamonds); αsantalene (mg l-1, filled squares); E,E-farnesol (mg l-1, filled
cycles) carbon dioxide production CTR (mmol h-1,lines) are
represented. Data represent the average of two independent
cultures.
how the culture metabolism was completely respiratory within the first 30 h characterized by
complete oxidation of glucose to biomass and carbon dioxide. This period was followed by a
phase where yeast growth was no longer consistent with the feeding profile producing a shift
towards fermentative metabolism accompanied by accumulation of glucose and ethanol and
characterized by a reduced sesquiterpene production rate. As consequence this result in a
suboptimal condition affecting productivity. The α-santalene final productivity obtained was 0.15
Cmmol gDCW-1 h-1.
3.4.2 Optimization of ISPR fed-batch process
During fed-batch operation the feed rate represents a key parameter in controlling the overall
process performances. From the result obtained in an initial stage of process optimization it was
clear that an operational strategy allowing optimal feeding policy was required to overcome the
problem related to the detrimental effect observed resulting from the overfeeding. Advances in
36
fermentation technology allow to design an optimal strategy to proper control the process.
Ideally a glucose based feed should proceed maintaining the sugar concentration below the
critical value preventing the metabolism switch to fermentative mode during the entire process.
A biomass coupled production process was designed applying RQ control together with the
previously developed glucose limited exponential feed. In the engineered strain, the extracellular
glucose concentration triggers the switch between the flux toward ergosterol biosynthesis and
product formation, and it is therefore critical to carefully control the cell metabolism to obtain the
best productivity. Instant RQ measurement was used as indicator of the cellular metabolic state
in order to fine-tune the feed profile to the cellular demand and maintain a fully respiratory state.
Figure 3.9. The configuration of the in situ product removal (ISPR) fed-batch RQ controlled
cultivation process. A stirred tank reactor is operated in fed-batch cultivation mode as double
phase system adding organic solvent on top of the culture and feeding concentrated culture
medium. The product is continuously captured in the organic phase due to its high hydrophobicity.
Feed delivery is designed with an exponential policy and it is controlled through a feed-back loop.
Fermentation exhaust gas analysis allows the on line determination of the respiratory quotient, the
instant RQ measurement modulated the feed addition by a PI controller in order to maintain the
desired set point.
The result of the respiratory quotient control on the feed profile is reported in figure 3.10. The
applied RQ feedback control method allows operating below the critical value alleviating the
glucose repression resulting in a robust feed strategy. Optimal feed policy resulted in the
maximum feed rate sustainable from the cell without byproduct formation. Extracellular
sesquiterpene analysis resulted in the detection of four different products, beside the target
compound α-santalene (representing 63 Cmol % of total sesquiterpene), the known side
product FOH, the minor product trans-α-bergamotene and E,E-farnesene. Apparently, in the
tested conditions SNS displayed different levels of specificity confirming a high degree of
plasticity that is often reported for this class of enzymes. This method resulted in a process
37
capable of producing 12 Cmmol l-1 (163 mg l-1) of α-santalene in 30 h with a product to
substrate yield of 0.0037 Cmmol Cmmol-1 and a productivity of 0.023 Cmmol gDCW-1 h-1.
Figure 3.10. Development of an RQ based feed-control ISPR aerobic glucose limited fed-batch cultivation.
Production phase of sesquiterpene producing S. cerevisiae strain SCIGS24. On-line measurement of CTR and
OTR allows instantaneous calculation of RQ value allowing to control the feed addition regulating the fed-batch
was performed using glucose with a concentration of 7.4 Cmol l (222 g l ) as a carbon source in the feed
media. Profiles reported represent the time course of (A) biomass formation (gDCW), RQ value (CTR/OTR),
comparison between exponential calculated feed profile and experimental value (ml) and (B) α-santalene, E,Efarnesol, trans-α-bergamotene and E,E-farnesol (Cmmol l ) product accumulation as function of feeding time.
Data presented are representative of three independent cultures; the error bars represent the standard deviation
for three independent cultivations
-1
-1
-1
3.4.3 Effect of ethanol as alternative carbon source to increase the precursor pool
The C2 carbon ethanol represents an attractive carbon source for secondary metabolite
production. However, because the raw material is often the dominating operative cost in
industrial bio-production of value-added chemicals (Otero et al., 2007) - although the market
price of ethanol decreased substantially due to the advent of bio-ethanol - a fully ethanol based
process would probably not be commercially valuable. Yeast S. cerevisiae has the ability to
simultaneously co-consume glucose-ethanol under fully respiratory conditions (Geurts et al.,
1980). The ratio between the two C-sources determines change in the metabolic flux in the
central carbon metabolism. If it is maintained below limits (0.57:0.43 Cmol Cmol-1) glucose
would be employed from the cell mainly for biosynthesis whereas all the cytosolyc acetyl-CoA
would be derived from ethanol that would then be used in the TCA cycle (de-Jong-Gubbels et
al., 1995; van Gulik et al., 1995). A mixed glucose/ethanol feed was applied to the previously
designed fed batch process. Under this condition sesquiterpenes are obtained directly from the
cytosolic acetyl-CoA produced from ethanol. Ethanol has a positive effect on process
performances yielding a 0.076 Cmmol gDCW-1 h-1 total sesquiterpene productivity equal to 49%
increase compared to the process with glucose alone and a final α-santalene titer of 14.2
38
Cmmol l-1 (193.4 mg l-1). Using this multiple carbon source strategy it was possible to use
ethanol as direct precursor for sesquiterpene and glucose for biomass intermediate
biosynthesis and increase the supply of acetyl-CoA precursor as a rapid approach to increase
the final product titer.
3.4.5 Double phase chemostat as tool for study metabolically engineered strains
Although fed-batch is by far the favored fermentation process in industrial production,
continuous cultivation modes find application in several industrial bioprocesses (e.g. insulin
production). Among the advantages offered by a chemostat system is the possibility to precisely
compare the productivities of selected genetically engineered strains under well controlled
constant growth condition and to explore the effect of specific growth rate independently of
other cultivations parameters. Combining chemostat cultivation with ISPR would be an effective
way to evaluate different strains and also a potential scaffold for an industrial process. This
system allows continuous recovery of the product in the fermenter effluent from the selected
organic phase which can subsequently be recycled, regenerated and reused in the same
process for prolonged time of cultivation (Fig 3.11).
Figure 3.11. Set-up of the in situ product removal (ISPR) chemostat cultivation process. A stirred
tank reactor is operated in continuous cultivation mode as double phase system feeding culture
medium and organic solvent. The product is continuously captured in the organic phase due to its
high hydrophobicity. In an integrated downstream step the two phases of the effluent are
partitioned in a settler. Subsequently, the product is recovered from the organic phase, which can
then be further recycled in the same process. The exhausted medium is discarded.
The effect of the specific growth rate that at these conditions is equal to the dilution rate (D) on
the sesquiterpene production of the engineered mutants was investigated in this study (Fig
39
3.12). Increasing the dilution rate (D=0.05 h-1 -> D=0.1 h-1), led to essentially unchanged
productivity and yield of the control strain (SCIGS28) suggesting that at higher dilution rates a
limitation of the plant synthase in efficiently draining the FPP precursor from the MVA pathway
when the FPP node was not manipulated to increase FPP availability. In the engineered strains
(SCIGS29 & SCIGS30) instead a slightly reduced α-santalene yield and a double productivity
was obtained corresponding to 0.041 and 0.043 Cmmol gDCW-1 h-1. This showed a clear
dependency of the productivity on the specific growth rate. The productivity level obtained
appears to be linearly correlated with the dilution rate employed pointing to a direct relation
between the specific growth rate and the flux through the MVA pathway .The same proportion
was maintained between α-santalene and the side product FOH under the two conditions
indicating that the distribution of the excess of FPP precursor between the two products
appears to be independent of the specific growth rate. This result suggests the hypothesis that
once a threshold level of intracellular flux toward FPP is reached the thermodynamically
favourable endogenous dephosphorylation starts to compete with the catalytic capacity of SNS
leading to FOH accumulation. On the other hand, the unchanged santalene yield coupled with
high productivity achieved at higher dilution rates suggests that SNS was not fully saturated at
low dilution rate and there was an excess activity to cope with higher FPP fluxes. This points out
that FOH formation is not only a direct consequence of limited SNS activity but other cellular
mechanisms are likely involved. Surprisingly, the highly engineered strains (SCIGS31, SCIGS24
& SCIGS25) were unable to grow at the higher specific growth rate applied. It is noteworthy that
comparable sesquiterpene productivity was achieved in the strain not fully engineered simply by
increasing the operational dilution rate whereas the fully engineered strains were washed out
when the same condition were imposed. Detailed analysis reveals that the introduced genetic
modifications profoundly affect the strain physiology, in particularly the carbon flux distribution
around the PDH bypass node, possibly affecting cytosolic acetyl-CoA availability and its
redistribution into the engineered MVA pathway. Further analyses are necessary to elucidate
factors leading to this particular physiological state that is most likely linked to the inability of the
fully engineered mutant to sustain growth at higher dilution rates.
40
Figure 3.12. Sequiterpene yield and productivities in a two-phase partitioned glucose-limited aerobic
chemostat. (A) α-santalene and farnesol production rate in Cmmol (g biomass)-1 h-1 of strains SCIGS28
(tHMG1↑), SCIGS29 (+ PHXT1-ERG9, lpp1Δ), SCIGS30 (+ dpp1Δ), SCIGS31 (+ gdh1Δ), SCIGS24 (+
ERG20↑, GDH2↑), SCIGS25 (+ upc2-1↑, tHMG1↑) cultivated at dilution rate D=0.05 h-1 and (B) strains
SCIGS28, SCIGS29 and SCIGS30 cultivated at dilution rate D=0.1 h-1. (C) α-santalene and farnesol
yield in Cmmol (Cmmol glucose) of strains SCIGS28, SCIGS29, SCIGS30, SCIGS31, SCOGS24 and
SCIGS25 cultivated at dilution rate D=0.05 h-1 and strain SCIGS28, SCIGS29 and SCIGS30 cultivated
at D=0.1 h-1. The C-molar weights of α-santalene and farnesol are, respectively, 13.62 and 14.82 g
Cmol-1. Error bars represent the standard deviation from three independent cultivations.
-1
3.5 Intracellular product accumulation and potential toxicity
The previously shown dependency of the productivity on the growth rate was reflected in the
inability to sustain a high dilution rate during chemostat cultivation and the impossibility to
maintain a constant growth rate during the fed-batch process. This affected the product
formation rate resulting in a suboptimal process. In order to investigate the possible causes of
the observed growth impediment the intracellular content of sequiterpene was assayed. Only
two of the four sesquiterpenes produced by the cell, α-santalene and α-trans-bergamotene,
were detected intracellularly with α-santalene as the main compound. Because cyclic terpenes
can accumulate in the membrane (Sikkema et al., 1994) it is reasonable to assume that this
structural difference might be the reason of the yeast capacity to specifically retain these
compounds inside the cell. The export of α-santalene from the cell seems to be the main cause
of the inhibitory effect that limits its productivity. Up to date the mechanism of secretion of
hydrocarbons are not known. The saturation kinetics on time observed for α-santalene might
suggest that yeast utilizes a specific transporter to secrete this compound. Transcriptome
analysis performed in presence of high sesquiterpene concentration was characterized by upregulation of multidrug transporters related genes in particular a pleiotropic drug resistant (PDR)
network was over-expressed, consistent with several parallel studies of yeast terpene producer
41
(Verwaal et al., 2010; Ro et al., 2008). Excessive α-santalene intracellular accumulation and an
intra-membrane derived toxicity related effect due to the loss of membrane integrity (Sikkema et
al., 1995) can explain the observed physiology.
Figure 3.13.
Typical extracellular (grey area) and intracellular (black area) sesquiterpene
accumulation profiles during RQ based double phase aerobic glucose or glucose/ethanol limited fed
batch cultivation of strain SCIGS24. (A) α-santalene; (B) trans-α-bergamotene. Typically the
intracellular fraction of α-santalene and trans-α-bergamotene were between 15-18% and 17-29% of
the total amount detected respectively. The two compounds displayed different intracellular
accumulation profile respect to the extracellular concentration; intracellular α-santalene reached
saturation around the value of 2 Cmmol . Error bars represent the standard deviation from three
independent cultivations.
-1
3.6 Expanding the substrate utilization range—toward a biorefinery
Techno-economical analysis of α-santalene production reveals how the raw material is often the
dominating operative cost and the design of a cost effective S. cerevisiae production platform
would rely on the accessibility to inexpensive carbon sources. Common industrial carbon
sources often consist of sugar mixtures (Olsson et al., 2000 Dahod et al., 2010). Lignocellulosic
biomass is among the most promising feedstocks that can provide sugar substrates for biobased production (Stephanopoulos, 2010). Xylose is one of the main components of
lignocellulosic feedstocks and is the second most abundant monosaccharide after glucose. Due
to the inability of wild type S. cerevisiae to efficiently utilize xylose as a sole carbon source, large
efforts have been invested form the scientific community to expand the substrate range
capability of S. cerevisiae (Van Vleet et al., 2009; Matsushika et al., 2009; van Maris et al., 2007;
Hahan-Hägerdal et al., 2007). Development of a cell factory for broader biomass-coupled
production would be favored without loss of carbon to overflow metabolites (ethanol, glycerol,
xylitol) particularly in the case of growth associated production processes like α-santalene
production that require simultaneous formation of biomass and target product. Here, S.
42
cerevisiae was metabolically engineered to consume xylose as an exclusive substrate
maximizing carbon flux to biomass production. Through the combination of (i) genetic
modifications (plasmid introduction) and (ii) application of selective pressure (shake flask
repetitive cultivations). (i) The oxido-reductive pathway of the native xylose-metabolizing yeast
Pichia stipitis consisting of three essential enzymes for xylose uptake (xylose reductase XR,
xylitol dehydrogenase XDH and xylulokinase XK) was reconstructed through plasmid-based
expression in S. cerevisiae. (Fig 3.14). The constructed mutant demonstrated the ability to
consume xylose aerobically although with a very poor growth.
Figure 3.14. Reconstruction of P. stipitis xylose assimilation pathway in S. cerevisae. (A) centromeric plasmid
pRS314-X123 for expression of XYL1 encoding xylose reductase (XR), XYL2 encoding xylitol dehydrogenase (XDH)
and XYL3 encoding xylulokinase (XK) all derived from P. stipitis cloned under the glyceraldehyde-3-phosphate
dehydrogenase (TDH3) constitutive promoter and terminator (Haiying et al., 2007). (B) Xylose utilization occurs in
three steps; first xylose is reduced to xylitol via XR (NADPH primarily consuming reaction), then xylitol is oxidized to
xylulose via XDH (NADH producing reaction), in the final conversion xylulose is phosphorylated to xylulose-5-P via XK
(ATP consuming reaction); xylulose-5-P is further chandelled into glycolytic intermediates via the pentose phosphate
pathway (PPP).
(ii) Successively, a directed evolution strategy was applied to select a spontaneous mutant with
improved xylose utilization rate using a repetitive batch cultivation technique. A mutant with
higher specific growth rate on xylose was selected by serial transfer of cells in batch shake flask
cultivation with minimal medium supplemented with xylose as the sole carbon source. This
approach targeted strain selection based on biomass formation rate directly coupled to xylose
consumption rate. A strain capable of rapid growth and fast aerobic xylose metabolism was
obtained in 21 days of selection period highlighting the efficiency and simplicity of the methods
and the high level of adaptability of S. cerevisiae when strong selective pressure is applied at
laboratory conditions (Fig 3.16). A total of 74 cell generations were used to obtain a 15-fold
increase in xylose consumption and a 52-fold increase in biomass production for the final
selected evolved strain The rapid adaptation observed confirm how early stage evolution plays a
critical role in the adaptation process and show how adaptation often occurs in few steps and
43
often involve only a limited number of mutations. The evolved strain exhibited a clearly
respiratory response toward xylose, its consumption was entirely oxidative with a high carbon
fraction converted to biomass (62% Cmol Cmol-1) and negligible amount of byproducts.
Figure 3.16. (A) Comparison in xylose consumption (bars) and biomass formation (line) during repetitive
growth of S. cerevisiae in shake flask (SF) cultures on synthetic medium with 20 g l-1 xylose. Shake flask
generation represents the number of specific shake flasks in series of repetitive cultivation performed to
select for mutants with higher specific growth rate and xylose utilization rates. The serial cultivation covers
10 cycles for a period of 500 h (21 days) after four batch cultures an improvement in xylose consumption
was detected. (B) Time course of aerobic batch culture on defined minimal medium supplemented with 20
g l xylose of the evolved strain, carbon evolution rate (CER) (dashed line), oxygen uptake rate (OUR) (solid
line) (mM h-1), and xylose (circle), ethanol (diamond) (g l-1); and biomass (square) (gDCW l-1) concentrations
as functions of cultivation time. Xylose was completely consumed within 60 h with a specific growth rate of
0.18 h-1 and biomass and carbon dioxide as the major fermentation product. Data represent the average of
three independent cultures.
-1
Transcriptional profiling was employed to further elucidate the observed physiology.
Transcriptome data support the physiological observation at the global and metabolic level.
Over-represented gene ontology (GO) process terms in the evolved strain were related to
function or features linked to respiratory processes. Transcription factor (TFs) enrichment
analysis identified factors primarily involved in carbon catabolite repression response including
transcriptional activators of genes involved in non-fermentative metabolism (Fig 3.17). Analysis
of gene expression at the metabolic pathway level reveals that a strong up-regulated glyoxylate
pathway plays an important role in enabling the observed respiratory metabolism. As extension
of the glyoxylate pathway up-regulation of cytosolic isocitrate dehydrogenase likely provides a
source of NADPH required to satisfy the biomass requirement. Xylose consumption in S.
cerevisiae through the oxido-reductive P. stipitis pathway has been often dominated by
extensive xylitol overflow ascribed as direct result of a redox imbalance of NAD(P) cofactors
between the XR and XDH and referred as the major drawback of the XR-XDH strategy.
Absence of xylitol formation in the evolved mutant under oxidative conditions may be interpreted
as a result of complete xylitol oxidation.
44
Figure 3.17. Integrated analysis of gene ontology (GO) process terms (A). “Glyoxylate and dicarboxylate
metabolism”, “peroxisome” and “oxidative phosphorylation” were among the most represented functional categories
with respect to metabolism of the evolved strain on xylose compared to un-evolved or evolved strain growing on
glucose. Transcription factor (TF) analysis (B). The main carbon catabolite repressor regulator SNF1 and several of
its known targets, the carbon source responsive ADR1 and the four subunits of the global respiratory regulator HAP
were the identified over-represented TFs in the evolved strain on xylose compared to un-evolved or evolved strain
growing on glucose. The cluster frequency is presented on the y-axis. Compared conditions are the evolved strain
cultivated on batch xylose (XEM_BX), evolved strain cultivated on batch glucose (XEM_BG) the unevolved strain
cultivated on batch glucose (WT_BG). Color key indicates the different expression in log-fold change (P <0.01).
adjusted
45
The result obtained in this study suggest that up-regulation of isocytrate deydrogenase ensures
sufficient NADPH production necessary for the P. stipitis NADPH-preferring XR to drive the
xylose catabolism, whereas the NADH surplus produced by XDH is reduced through the
respiration eliminating the NADP+/NAD+ imbalance.
CHAPTER 4 Conclusions & Future Prospects
4.1 Conclusions
The main objective of this research
was the construction of an efficient S.
cerevisiae
capable
of
producing industrially relevant titers of
!"!(!#
)*+,*-.+.##/01234,565,7##89::1-;<=9>#?@#
cell factory
the
!"!'!#
sequiterpene
α-santalene,
a
precursor of commercially interesting
compounds. A
!"!&!#
rationally designed
metabolic engineering approach was
!"!%!#
employed to addresses the feasibility
!"!$!#
of
sesquiterpene
production
!"!!!#
A//#E0*+4?#/15+,#
BC,5:5D*,51+#
in
sesquiterpene
FGH#C*,?I*7#
BC,5:5D*,51+#
A.0:.+,*,51+#
BC,5:5D*,51+#
Figure 4.1 α-santalene productivity progression (Cmmol
gDCW-1 h-1) achieved during this study compared to the
reference strain minimally engineered (Chapter 3.1.1),
applying different strategies: FPP branch point optimization
(Chapter 3.2) MVA optimization (Chapter 3.3) and
fermentation optimization (Chapter 3.4)
S.
α-santalene
cerevisiae.
production
The
was
optimized using a novel strategy to
dynamically modulate gene expression
around the key FPP branch point.
Using this strategy it was possible to
divert carbon flux toward the desired
compound. Subsequently, a multistep
metabolic engineering approach to
increase precursor and cofactor supply was used to further optimize isoprenoid accumulation.
This involved introducing genetic modifications that enabled channeling an increased flux
through the isoprenoid pathway. This was combined with fermentation optimization and an
integrated product recovery process developed improving yield, productivity and titer (Fig 4.1).
Finally, the substrate range of the producing platform was expanded through a complementary
evolutionary engineering approach to obtain a cell factory capable of exclusively consuming
xylose as a carbon source. The results presented in this study represent a gateway towards the
46
creation of an industrial microbial platform that can be applied to the production of an array of
sesquiterpene products.
4.2 Perspectives
The future feedstock of chemical manufacturing will not be oil but glucose. As we are
approaching the “artificial biotechnology era” (Gibson et al., 2010; Gibson 2008; Lartigue et al.,
2009), one can imagine a future where specialized ”microrefineries” will produce a portfolio of
renewable chemical compounds from waste raw material, accomplishing the society the
demand to produce via a responsible and sustainable way and hereby reduce the production
cost.
Our society is reaching a turning point; an intriguing question to ask for the future is if the
creation of custom build-cell factories for the production of a wide spectrum of strategic
chemicals would be a feasible realistic goal in a near future. The answer is not simple and a
number of challenges lie ahead. The continuous advance in metabolic engineering and its
complementary related disciplines allow a high degree of manipulation on model industrial
microorganism producing spectacular examples of successful biological production of needed
chemicals (Westfall et al., 2011; Ajikumar et al., 2010). However, the ability to rapidly engineer
microbial cells is still in its infancy and most of their immense potential remains unexplored. The
time consuming development of a toolbox of genetic components to control gene expression
has slowed down the capacity to manipulate cell features, due to technology limitations and the
unpredictable complexity of the cell system. Although the scientific community strives to
develop open technology platforms for designing and cataloging biological parts (Hayes et al.
2001) that culminated in the creation of registries of “standard” biological parts for synthetic
biology, like for example “BioFAB” (Biofab.org) and the “Registry of Standard biological Parts”
(partsregistry.org) collecting several thousand of genetic elements (Baker et al., 2006), most of
these parts are still uncharacterized or incompatible and unreliable (Kwok, 2010; Kelly et al.,
2009) resulting mainly in host-dependent application and therefore difficult to use universally.
Additionally, several of the essential design principles for engineering microbes are still missing
(Morton 2005; Knight 2005). Similarly, it would be necessary to acquire the same high level of
standardization for mathematical design of biological models, in order to empower accuracy
and robustness in the predictive capacity of modeling; a notable example in the effort to uniform
the rules of model language is the Systems Biology Markup Language (SMBL) (Hucka et al.,
2003).
As every rapidly moving area the current efforts in microorganism manipulation could rise a
number of ethical, safety, legal and security issues (not discussed here) that could limit near
47
term commercial opportunities and compromise their marketing success (Dana et al., 2012; Tait
et al., 2012; Erickson et al., 2011; Palombi 2009, Venter et al., 2004). Nevertheless, steady
progress has been made in pursuing the formidable challenge of bio-manufacturing
compounds. The time is ripe, the field is clearly moving from the discovery phase to the
application and implementations, and the achievements accomplished so far are encouraging,
even though it is clear that major efforts are still required before biological engineering to provide
tailored products will became routine.
The current approaches mainly rely on optimization or transfer of a specific activity or an existing
pathway between the natural producers selected production organisms, this approach is limited
by the number of modification that the engineered strain can tolerate. However, the limits in the
creation of custom-build cell factories are only temporary. In the future, the scale down in the
price of gene synthesis (today the cost for DNA fragment is less than $0.1/Mb) (Wetterstrand,
2012) would allow to bypass the traditional time consuming DNA manipulation routinely
replacing it with automated synthesis (Carr et al., 2009). The increased effort from the scientific
community to develop engineered toolsets and cataloged parts (Baker et al., 2006) would allow
to fast design and construct cells with new desired characteristics and features, resulting in a
more standardized and automated strain engineering process that would lead to a more
systematic practice of metabolic engineering rather than the specific case-by-case engineering
which we are nowadays assisting. Inspired by the natural cooperative of microbial population
recent advances in cell-to-cell communication (Weiss et al., 2005; You et al., 2004) bright light
to the opportunity to engineer microbial consortia rather than single organism. Albeit in its young
phase co-culture engineering may help in performing multi-task processes in a more efficient
manner than a single organism culture (Brenner et al., 2008).
To conclude it is worth to mention how the pioneering work in genome engineering conducted
at the J. Craig Venter institute (Rockville, MD, USA) that recently lead to the extraordinary
breakthrough in synthetic engineering (Gibson et al., 2010), paved the way to a radical new
approach consisting of a de novo artificially assembled cell with an entirely in silico designed
and synthetically synthesized genome that could be programmed to produce fuels, chemicals
and medicine. Another interesting approach holding the promise to impact in the future the
chemical industry is the use of cell-free systems. Albeit still in its early stage this technology
could offer a flexible way to produce chemicals without using intact cells, and bypassing the
constraints related to the cell wall (Hodgman et al., 2012). These inspirational goals could
reshape in a long term prospective the metabolic engineering field.
48
While challenges remain, metabolic engineering has and would continue to have a tremendous
impact in our society; the stage for the green technology era based on engineered biosystems
is set.
49
ACKNOW LEDGEMENTS
When I started my doctorate I felt like I was 1000 miles away from where I wanted to be, I knew
that if I would have chosen the wrong way that would have dig hole in my way, put rocks in my
way and make the 1000 miles harder to endure, luckily on this path I was not alone but
surrounded by amazing people and certainly, this dissertation would not come possible without
their help, contribution and dedication.
First and foremost, I offer my sincerest gratitude to my supervisor Professor Jens Nielsen. Jens I
remember the day we first met in your office in Denmark and you started writing pathway
engineering strategies for me in a small piece of paper, by that very moment everything became
clear and I realized what I would have done in my future. Well, I still jealously preserve that pice
of paper that changed the next 4 years of my life. It is very difficult to express my gratitude with
words, throughout my PhD path you inspire me with outstanding guidance, positivity and
brilliant science, constantly stimulating my curiosity and intuition, and perhaps most importantly
your trust and unlimited support. I am forever indebted to Dr. Jose Manuel Otero; Manny
meeting you was a life changing event you transmit me your passion and enthusiasm for
science and give me the confidence to follow my heart and find what I love to do. I hope one
day I will have the chance and the strength to repay what you have done for me as scientist and
friend. I am grateful to Dr. Verena Siewers for her patient assistance and supervision. I would
like to extend my deepest gratitude to Siavash Partow, this dissertation would not have been
possible unless your help, support, brotherhood and friendship. I would like to show my
gratitude to my colleagues and collaborator that have been part of the ”Isoprenoids project”, Dr.
Laurent Daviet, Dr. Michel Schalk, Dr. Jerome Maury and Dr. Yun Chen. I am grateful to the
Post-doctoral associates Dr. Intawat Nokaew, Dr. Rahul Kumar, Dr. Nandy Kumar, Dr. Kim IlKwon, and Dr. Luis Caspeta, and fellow PhD students Kanokarn Kocharin, Kuk-ki Hong, Juan
Octavio Valle, Leif Varemo and Tobias Osterlund for the stimulating scientific discussions and
creative atmosphere. I also would like to thank Dr. Valeria Mapelli, Dr. Jie Zhang, Dr. Wanwipa
Vongsangnak, Dr. Roberto Olivarez, Dr. Margarita Salazar, Dr. Pramote Chumnanpuen, the
core group of close friends that moved from DTU to start with me the new adventure at
Chalmers. I am thankful to my student Christoph Knuf, for giving me the opportunity to be your
teacher and to share knowledge with you. I wish to say a tremendous thank to Professor
Lisbeth Olsson for her support and education that help me to maturate when I was young
researcher at DTU. I also would like to express my appreciation to Professor Goutham Vemuri
for the valuable scientific input during the early stage of my PhD career. A special thanks goes
to Professor Dina Petranovic for her precious support during the transition phase at the end of
50
my PhD. My acknowledgement also goes to the administrative staff Erica Dahlin and Martina
Butorac and the technical staff Malin Nordvall, Ximena Sevilla, Pegah Khorramzadeh and
Suwanee Jansa-Ard. There are many more that deserve to be listed here; I take this opportunity
to thank all those who have contributed in any way to my PhD study. This PhD study was
funded by Chalmers Foundation (Sweden), Firmenich SA (Switzerland), and the Knut and Alice
Wallenberg Foundation (Sweden).
From the depth of my heart I am eternally grateful to my family Mam, Dad, big Brother and little
Sister, for their constant love and unwavering support. Finally, my the best thanks goes to the
most important person of my life, Marta, you above all other people understand the difficulties
that accomplish this doctorate can represent, deciding to undertake this path together and
sharing our life through this amazing experience you allow me to be at my strongest, you are my
inspiration, my love and my life. Ultimately, big thanks go to all the friends at GA that help me to
focus during the end of my PhD.
Gionata Scalcinati
June, 2012
51
REFERENCES
Ajikumar PK, Xiao WH, Tyo KE, Wang Y, Simeon F, Leonard E, Mucha O, Phon TH, Pfeifer B, Stephanopoulos G.
2010. Isoprenoid pathway optimization for taxol precursors overproduction in Escherichia coli. Science 330:70-74.
Albertsen L, Chen Y, Bach LS, Rattleff S, Maury J, Brix S, Nielsen J, Mortensen UH. 2001. Diversion of flux toward
sesquiterpene production in Saccharomyces cerevisiae by fusion of host and heterologous enzymes. Appl Environ
Microbiol 77:1033-1040.
Alper H, Stephanopoulos G. Engineering for biofuels: exploiting innate microbial capacity or importing biosynthetic
potential? 2009. Nature Rev. 7:715-723.
Alper H, Moxley J, Nevoigt E, Stephanopoulos G. 2006. Engineering yeast transcription machinery for improved
ethanol tolerance and production. Science 314:1565-1568.
Alper H, Fisher C, Nevoigt E, Stephanopoulos G. 2005. Tuning genetic control through promoter engineering. Proc
Natl Acad Sci USA 102:12678-12683.
Asadollahi MA, Maury J, Schalk M, Clark A, Nielsen J. 2010. Enhancement of farnesyl diphosphate pool as direct
precursor of sesquiterpenes through metabolic engineering of the mevalonate pathway in Saccharomyces cerevisiae.
Biotechnol Bioeng 106:86-96.
Asadollahi M, Maury J, Patil KR, Schalk M, Clark A, Nielsen J. 2009. Enhancing sesquiterpene production in
Saccharomyces cerevisiae through in silico driven metabolic engineering. Metab Eng 11: 328-334.
Asadollahi MA, Maury J, Møller K, Nielsen KF, Schalk M, Clark A, Nielsen J. 2008. Production of plant sesquiterpenes
in Saccharomyces cerevisiae: Effect of ERG9 repression on sesquiterpene biosynthesis. Biotechnol Bioeng 99:666677.
Auslander S, Auslander D, Muller M, Wieland M, Fussenegger M. 2012. Programmable single cell-cell mammalian
biocomputer. Nature doi:10.1038/nature11149.
Bailey JE. 1991. Toward a science of metabolic engineering. Science 252:1668-1674.
Bayer TS, Smolke CD. 2005. Programmable ligand-controlled riboregulators of eukaryotic gene expression. Nat
Biotechnol.23:337-343.
Baker D. Chrch G, Collins J, Endy D, Jacobson J, Keasling J, Modrich P, Smolke C, Weiss R. 2006. Engineering life:
building a fab for biology. Sci Am 294:44-51.
Braatsh S, Helmark S, Kranz H, Koebmann B, Jansen PR. 2008. Escherichia coli strains with promoter libraries
constructed by Red/ET recombination pave the way for transcriptional fine-tuning. BioTechnique 45:335-337.
Barrick JE, Yu DS, Yoon SH, Jeong H, Oh TK, Schneider D, Lenski RE, Kim J. 2009. Genome evolution and
adaptation in a long-term experiment with Escherichia coli. Nature 461:1243-1247.
Basu S, Gerchman Y, Collins CH, Arnold FH, Weiss R. 2005. A synthetic multicellular system for programmed
pattern formation. Nature 434:1130-1134.
Basson ME, Thorsness M, Rine J. 1986. Saccharomyces cerevisiae contains two functional genes encoding 3hydroxy-3-methylglutaryl-coenzyme A reductase. Proc Natl Acad Sci USA 83:5563-5567.
Benecia F, Courreges. 1999. Antiviral activity of sandalwood oil against Herpes simplex viruses-1 and -2.
Phytomedicine 6:119-123.
Blom J, De Mattos T. Grivell LA. 2000. Redirectionof the respire-fermentative flux distribution in Saccharomyces
cerevisiae by overexpression of the transcription factor Hap4p. Appl Environ Microbiol 66:1970-1973.
Blount BA, Weenink T, Vasylechko S, Ellis T. 2012. Rational diversification of a promoter providing fine-tuning
expression and orthogonal regulation for synthetic biology. PLOS One 7:e33279.
Bohlmann J, Keeling CI. 2008. Harnessing plant biomass for biofuel and biomaterial. Terpenoid biomaterials. Plant J.
54:656-669.
Bohlman GM, 2005.Biorefinery process economics. Chem Eng Process 10:37-43.
52
Bonoli M, Graziola M, Poggi V, Hochkoeppler A.2006. RNA complementary to the 5’ UTR of mRNA triggers effective
silencing in Saccharomyces cereviaise. Biochem Biophys Res Commun 339:1224-1231.
Bourque JE. 2005. Antisense strategies for genetic manipulations in plant. Plant Sci 105:125-149.
Boyle PM, Silver PA. 2009. Harnessing nature’s toolbox: regulatory elements for synthetic biology. J R Soc Interface
6:S535-S546.
Brenner K, You L, Arnold FH. 2008. Engineering microbial consortia: a new frontier in synthetic biology. Cell 26:483489
Brown MS, Goldstein JL. 1980. Multivalent feedback regulation of HMG CoA reductase, a control mechanism
coordinating isoprenoid synthesis and cell growth. J Lipid Res 21:505-517.
Brunke EJ, Schmaus G. 1995. New odor-active constituents in sandalwood oil. Dragoco Report 42:197-245
Burgard AP, Pharkya P, Maranas CD. 2003. Optknock: A bilevel programming framework for identifying gene
knockout strategies for microbial strain optimization. Biotechnol Bioeng 84:647-657.
Cakar ZP, Turanli-Yildiz B, Alkim C, Yilmaz U. 2012. Evolutionary engineering of Saccharomyces cerevisiae for
improved industrially important proprieties. FEMS Yeast Res. 12:171-182.
Canton B, Labno A, Endy D. 2008. Refinement and standardization of synthetic biological parts and devices. Nat
Biotechnol 26:787-793.
Carr PA, Church GM. 2009. Genome engineering. Nat Biotechnol 12:1151-1162.
Chapuis C. 2004. In the quest for virtual pseudo receptor for sandalwood-like odorants. Chemistry & Biodiversity
1:980-1021.
Chambon C, Ladeveze V, Oulmouden A, Servouse M, Karst F. 1990. Isolation and properties of yeast mutants
affected in farnesyl diphosphate synthetase. Curr Genet 18: 41-46.
Chatterjee R, Yuan L. 2006. Direct evolution of metabolic pathway. Trends in Biotechnology 24:28-38.
Chen Y, Partow S, Scalcinati G, Sewers V, Nielsen J. 2012. Enhancing the copy number of episomal plasmids in
Saccharomyces cervisiae for improved protein production. FEMS Yeast Research. DOI: 10.1111/j.15671364.2012.00809.x
Cho H, Cronan JE. 1995. Defective export of a periplasmic enzyme disrupts regulation of fatty acid synthesis. J Biol
Chem. 270:4216:4219.
Christianson DW. 2008. Unearthing the roots of the terpenome. Curr Opin Chem Biol 12:141-150.
Craig L. 2012. U.S. founds biofuels for military, transport. Biofuels, renewable energy, Earth techling newsletter.
Da Silva NA, Srikrishnan S. 2012. Induction and expression of genes for metabolic engineering application in
Saccharomyces cerevisiae. FEMS Yeast Res 12:197-214.
Dana GV, Kuiken T, Rejeski D, Snow AA. 2012. Synthetic bioogy: four steps to avoid a synthetic biology disaster.
Nature 483: 29
Danishefsku Y SJ, Masters JJ, Young WB, Snydder LB, Magee TV, Jung DK, Isaacs RC, Bornmann WG, Alaimo CA,
Coburn CA, Di Grandi M. 1996. Total synthesis of baccantin III and taxol. J Am Chem Soc 118:2843-2859.
Daum G, Tuller G, Nemec T, Hrastnik C, Balliano G, Cattel l, Milla P, Rocco F, Conzelmann A, Vionett C, Kelly DE,
Kelly S, Schweizer E, SChuller HJ, Hojad U, Greiner E, Finger K. 1999. Yeast functional analysis reports. Systematic
analysis of yeast strains with possible defects in lipid metabolism. Yeast 15:601-614.
Davies E. 2009. The sweet scent of success, Fragrance Chemistry. Chemistry world 40 February 2009.
Davies BSJ, Wang HS, Rine J. 2005. Dual activators of the sterol biosynthetic pathway of Saccharomyces cerevisiae:
similar activation/regulatory domains but different response mechanisms. Mol Cell Biol 25: 7375-7385.
De-Jong-Gubbels P, vanrolleghem P, Heijnen S, van Dijken JP, Pronk JT. 1995. Regulation of carbon metabolism in
chemostat cultures of Saccharomyces cerevisiae growth on mixtures of glucose and ethanol. Yeast 11:407-418.
53
DeJong JM, Liu Y, Bollon AP, Long RM, Jennewein S, Williams D, Croteau RB. 2005. Genetic engineering of taxol
biosynthetic genes in Saccharomyces cerevisiae. Biotechnol Bioeng 93:212-224.
Dellomonaco C, Fava F, Gonzales R. 2010. The path to next generation biofuels: successes and challenges in the
era of synthetic biology. Microb Cell Fact 9:3.
Dhaod SK, Greasham R, Kennedy M. Raw material selection and medium development for industrial fermentation
processes. 2010. Baltz R, Davies JE, DEmian AL Eds. Manual of industrial microbiology and biotechnology. ASM
Press, Washington, DC, USA pp.659-668.
Donald KA, Hampton RY, Fritz IB. 1997. Effects of overproduction of the catalytic domain of 3-hydroxy-3methylglutaryl coenzyme A reductase on squalene synthesis in Saccharomyces cerevisiae. Appl Environ Microbiol
63:3341-3344.
Dornburg V, Hermann BG, Patel MK. S2008. Scenario projections for future market potential of biobased bulk
chemicals. Environ Sci Technol 42:2261-2267.
dos Santos M, Thygesen G, Kötter P, Olsson L, Nielsen J. 2003. Aerobic physiology of redox-engineered
Saccharomyces cerevisiae strains modified in the ammonium assimilation for increased NADPH availability. FEMS
Yeast Res 4: 59-68.
Dueber JE, Wu GC, Malmirchegini GR, Moon TS, Petzold CJ, Ullal AV, Parther KLJ, Keasling JD. 2009. Synthetic
protein scaffolds provide modular control over metabolic flux. Nat Biotechnol 27:753-759.
Dulaney EL, Stapley EO, Simpf K. 1954. Studies on ergosterol production by yeast. Appl Microbiol 2:371–379.
Dwivedi, C., Guan, X., Harmsen, W.L., Voss, A.L., Goetz-Parten, D.E., Koopman, E.M., Johnson, K.M., Valluri, H.B.,
Matthees, D.P., 2003. Chemopreventive effects of alpha-santalol on skin tumor development in CD-1 and SENCAR
mice. Cancer Epidemiol Biomarkers Prev 12:151-156.
Endy D. Foundations for engineering biology. Nature 438:449-453.
Enriquez J. 2009. What matters: how biotech will reshape the global economy. McKinsley&Company.
Enyeart PJ, Ellington AD. 2011. A yeast for all reasons. Nature 447:413-414.
Erickson B, Singh R, Winters P. 2011. Synthetic biology: regulating industry uses of new biotechnologies. Science
333:1254-1256.
FAO Food and Agriculture Organization of the United Nations corporate document repository. 1995 Flavor and
Fragrances of plant origin, Chapter 6, Sandalwood oil. ISBN 92-5-103648-9.
Fang F, Salmon K, Shen MW, Aeling KA, Ito E, Irwin B, Tran UC, Hatfield GW, Da Silva NA, Sandmeyer S. 2011. A
vector set for systematic metabolic engineering in Saccharomyces cerevisiae. Yeast 28:123-136.
Farhi M, Marhevka E, Masci T, Marcos E, Eyal Y, Ovadis M, Abeliovich H, Vainstein A. 2011. Harnessing yeast
subcellular compartments for the production of plant terpenoids. Metab Eng 13: 474-481.
Farmer WR, Liao JC. 2000. Improving lycopene production in Escherichia coli by engineering metabolic control. Nat
Biotechnol. 18:533-537.
Festel G. 2010. Industrial biotechnology: market size, company types, business models, and growth strategies. Ind
Biotecnol 6:88-94.
Flagfeldt DB, Siewers V, Huang L, Nielsen J. 2009. Characterization of chromosomal integration sites for
heterologous gene expression in Saccharomyces cerevisiae. Yeast 26: 545-551.
Fortman JL, Chhabra S, Mukhopadhyay A, Chou H, Lee TS, Steen E, Keasling JD. 2008. Biofuel alternatives to
ethanol: pumping the microbial well. Cell 26: 375-381.
Frick O, Wittmann C. 2005. Characterization of metabolic shift between oxidative and fermentative growth in
Saccharomyces cerevisiae by comparative C flux analysis. Microb Cell Fact 4:30.
13
GEN News. 2008. OneWorld healt, Amyris, and Sanofi-Aventis to create a complementary source of key ingredient of
malaria therapy. Genetic Engineering & Biotechnology News Highlight, Mar3, 2008.
54
GEN News. 2010. Tate & Lyle to manufacture farnesene for Amyris. Genetic Engineering & Biotechnology News
Highlight, Nov 5, 2010.
Geurts TG, de Kok HE, Roles JA. 1980. A quantitative description of the growth of Saccharomyces cerevisiae CBS
426 on amixed substrate of glucose and ethanol. Biotechnol Bioeng 22: 2031-2043.
Gibson DG, Glass JI, Lartigue C, Noskov VN, Chuang RY, Algire MA, Benders GA, Montague MG, Ma L, Moodie
MM, Merryman C, Vashee S, Krishnakumar R, Assad-Garcia N, Andrews-Pfannkoch C, Denisova EA, Young L, Qi
ZQ, Segall-Shapiro TH, Calvey CH, Parmar PP, Hutchinson III CA, Smith HO, Venter CJ. 2010. Creation of a bacterial
cell controlled by a chemically synthesized genome. Science 329:52-56.
Gibson DG, Benders GA, Andrews-Pfannkoch C, Denisova EA, Baden-Tillson H, Zaveri J, Stockwell TB, Brownley A,
Thmoas DW, Algire MA, Merryman C, Young L, Noskow VN, Glass J, Venter CJ, Hutchinson III CA, Smith HO. 2008.
Complete chemical synthesis, assembly, and cloning of a Mycoplasma genitalium genome. Science 319:1215.1220.
Good L. 2003. Translational repression by antisense sequences. Cell Mol Life Sci 60:854-861.
Grabinska K, Palamarczyk G. 2002. Dolichol biosynthesis in the yeast Saccharomyces cerevisiae: an insight into the
regulatory role of farnesyl diphosphate synthase. FEMS Yeast Res 2: 259-265.
Gustafsson C, Govindarajan S, Minshull J. 2004. Codon bias and heterologous protein expression. Trends
Biotechnol 22:346-353.
Hampton R, Dimster-Denk D, Rine J. 1996. The biology of HMG-CoA reductase: the pros of contra-regulation.
Trends Biochem Sci 21:140-145.
Hampton RY, Rine J. 1994. Regulated degradation of HMG-CoA reductase, an integral membrane protein of the
endoplasmic reticulum, in yeast. J Cell Biol 125:299-312.
Hahn-Hägerdal B, Karhumaa K, Jeppson M, Gorwa-Grausland MF. 2007. Metabolic engineering for pentose
utilization in Saccharomyces cerevisiae. Adv Biochem Eng Biotechnol 108: 147-177.
Haiying N, Laplaza JM, Jeffries TW. 2007. Transposon mutagenesis to improve the growth of recombinant
Saccharomyces cerevisiae on D-Xylose. Appl Environ Microbiol 73: 2061-2066.
Hayes B. 2001. Computing comes to life: How to build a computer out of E. Coli. Am Sci 89:204.
Heijnen JJ, van Scheltinga TA, Straathof AJ.1992. Fundamental bottlenecks in the application of continuous
bioprocess. J Biotechnol. 22: 3-20.
Hodgmann CE, Jewett MC. 2012. Cell-free synthetic biology: thinking outside the cell. Metab Eng 14:261-269.
Holtz WJ, Keasling JD. 2010. Engineering static and Dynamic control of Synthetic pathways. Cell 140:19-23.
Hong K, Nielsen J. 2012. Metabolic engineering of Saccharomucse cerevisiae a key cell factory platform for future
biorefineries. Cell Mol Life Sci. in press.
Hou J, Scalcinati G, Oldiges M, Vemuri GN. 2010. Metabolic impact of increased NADH availability in
Saccharomyces cerevisiae. Appl Environ Microbiol 76: 851-859.
Howes MJ, Simmonds MS, Kite GC. 2004. Evaluation of the quality of the sandalwood essential oils by gas
chromatography-mass spectrometry. J Chromatogr A. in press.
Hucka M, Finney A, Sauro HM,Bolouri H, Doyle JC, Kitano H.2003 The systems biology markup language (SBML): A
medium for representation and exchange of biochemical network models. Bioinformatics 19:524-531.
Isaacs FI, Dwyer DJ, Ding C, Pervouchine DD, Cantor CR, Collins JJ. 2004. Engineerign riboregulators enable posttranscriptional control of gene expression. Nature Biotechnol 22:841-847.
Jackson BE, Hart-Wells EA, Matsuda SP. 2003. Metabolic engineering to produce sesquiterpenes in yeast. Org Lett
5: 1629-1632.
Jansen LA, Diderich AJ, Mashego M, Hassane A, de Winde HJ, Daran-Lapujade P, Pronk TJ. 2005. Prolonged
selection in aerobic, glucose-limited chemostat cultures of Saccharomyces cerevisiae causes partial loss of glycolitic
capacity. Microbiology 151: 1657-1669.
55
Jewett MC, Oliveira AP, Patil KR, Nielsen J. 2005. The role of high-throughput transcriptome analysis in metabolic
engineering. Biotechnol Bioprocess Eng 10:385-399
Jones CG, Moniodis J, Zulak KG, Scaffidi A, Plummer JA, Ghisalberti EL, Barbour EL, Bohlmann J. 2011.
Sandalwood fragrance biosynthesis involves sesquiterpene synthases of both the terpene synthase (TPS) and TPS-b
subfamily, including santalene synthases. J Biol Chem. In press.
Jirovetz L, Buchbauer G, Denkova Z, Stoyanova A, Murgov I, Gearon V, Birkbeck S, Schmidt E, and Geissler M.
Comparative study on the antimicrobial activities of the different sandalwood essential oils of various origin. Flavour
Fragr J 21:465-468.
Kamm B, Kamm M.2004. Principles of biorfineries. Appl Microbiol Biotechnol 64: 137-145.
Kaufmann A, Knop M. Genomic replacement cassettes to alter gene expression in yeast Saccharomyces cerevisiae.
Methods Mol Biol 765:275-294.
Katie N. 2010. Soliance partners with Amyris to produce renewable squalane. Cosmetics 23 Jun, 2010.
Keasling .2012. Synthetic biology and the development of tools for metabolic engineering. Metab Eng in press.
Keasling JD. 2010. Manufacturing molecules through metabolic engineering. Science 330:1355-1358.
Keasling JD. 2007. Syntetic biology for synthetic chemistry. ACS Chem Biol 3:64-76.
Keasling JD. 1999. Gene expression tools for the metabolic engineering of bacteria. Tibitech 17:452-460.
Keith ET, Alper HS, Stephanopoulos G. 2007. Expanding the metabolic engineering toolbox: more options to
engineering cells. Trends Biotechnol 25:132-137.
KellyJR, Rubin AJ, Davbis JH, Ajo-Frankling MC, Cumbers J, Czar MJ, de Mora K, Glieberman AL, Monie DD, Endy
D. 2009. Measuring the activity of BioBrick promoters using an in vivo reference standard. J Biol Eng 3:4
Kennedy, M.A., Barbuch, R., Bard, M. 1999. Transcriptional regulation of the squalene synthase gene (ERG9) in the
yeast Saccharomyces cerevisiae. Biochim. Biophys. Acta 1445: 110-122.
Kennedy, M.A. 2001. Positive and negative regulation of squalene synthase (ERG9) and ergosterol biosynthetic gene,
in Saccharomyces cerevisiae. Biochem. Biophys. Acta 1517: 177-189.
Khalil AS, Collins JJ. 2010. Synthetic biology: applications come of age. Nat Rev Gen 11: 367-379.
Kim IK, Roldao A, Siewers V, Nielsen J. 2012. A systems-level approach for metabolic engineering of yeast cell
factories. FEMS Yeast Res 12:228-248.
Kirby J, Romanini DW, Paradise EM, Keasling JD. 2008. Engineering triterpene production in Saccharomyces
cerevisiae - beta-amyrin synthase from Artemisia annua. FEBS J 275:1852-1859.
Knight TF. 2005. Engineering novel life. Mol Syst Biol 1:2005:0020.
Koepp AE, Hezari M, Zajicek J, Vogel BS, LaFever RE, Lewis NG, Croteau R. 1995. Cyclization of geranylgeranyl
diphosphate to taxa-4(5),11(12)-diene is the committed step of taxol biosynthesis in pacific yew. J Biol Chem
270:8686-8690.
Kuyper M, Toirkens JM, Diderich AJ, Winkler AA, van Dijken PJ, Pronk TJ. 2005. Evolutionary engineering of mixedsugar utilization by a xylose-fermenting Saccharomyces cerevisiae strain. FEMS Yeast Res 5: 925-934.
Kuzuyama T, Seto H. 2003. Diversity of biosynthesis of the isoprene units. Nat Prot Rep 20:171-183.
Kwok R. 2010. Five hard truths for synthetic biology, can engineering approach tame the complexity of living
systems? Nature 463:288-290.
Lange BM, Rujan T, Martin W, Croteau R. 2000. Isoprenoid biosynthesis: the evolution of two ancient and distinct
pathway across genomes. Proc Nat Acad Sci 97:13172-13177.
Lartigue C, Vashee S, Algire MA, Chuang RY, Benders GA, Ma L, Noskov VN, Denisova EA, Gibson DG, AssadGarcia N. Alperovich N, Thomas DW, Merryman C, Hutchinson III CA, Smith HO, Venter JC, Glass JI. 2009. Creating
bacterial strain from genomes that have been cloned and engineered in yeast. Science 325:1693-1696.
56
Lee JW, Kim TY, Jang YS, Choi S, Lee SY. 2011 . Systems metabolic engineering for chemicals and materials. Cell
29:370-378.
a
Lee TS, Krupa RA, Zhang F, Hajimorad M, Hltz W, Prasad N, Lee SK, Keasling JD. 2011 . BglBrick vectors and
datasheets: a synthetic biology platform for gene expression. J Biol Eng 5:12
b
Lee H, DeLoache W, Deuber JE. 2011 . Spatial organization of enzymes for metabolic engineering. Metab Eng
14:242-251.
c
Lee SK, Chou H, Ham TS, Lee TS, Keasling JD. 2008. Metabolic engineering of microorganisms for biofuel
production: from bugs to synthetic biology to fuels. Curr Opin Biotechnol 19:556-563.
Leib TM, Pereira CJ, Villadsen J. 2001. Bioreactors: a chemical engineering prospective. Chem Eng Sci 56:54855497.
Leonard E, Ajikumar PK, Thayer K, Xiao WH, Mo JD, Tidor B, Stephanopoulos G, Parther KL. 2010. Combining
metabolic and protein engineering of a terpenoid biosynthetic pathway for overproduction and selectivity control.
Proc Natl Acad Sci 107:13654-13659.
Lewis TL, Keesler AG, Fenner GP, Parks W. 1988. Pleiotropic mutations in Saccharomyces cerevisiae affecting sterol
uptake and metabolism. Yeast 4: 93-106.
Lombard J, Moreira D. Origins and evolution of the mevalonate pathway of isoprenoids biosynthesis in the three
domain of life. Mol Biol Evol 28:87-99.
Madsen KM, Udatha GD, Semba S, Otero JM, Koetter P, Nielsen J, Ebizuka Y, Kushiro T, Panagiotou G. 2011.
Linking genotype and phenotype of Saccharomyces cerevisiae strains reveals metabolic engineering targets and
leads to triterpene hyper-producers. Plos One 6: e14763.
May M. 2009. Engineering a new business. Nature Biotechnol 12:1112-1120.
Martin VJ, Pitera DJ,Withers S, Newman JD, Keasling JD. 2003. Engineering a mevalonate pathway in Escherichia
coli for production of terpenoids. Nature Biotechnol 21:796-802
Martin VJ, Yoshikuni Y, Keasling JD. 2001. The in vivo synthesis of plant sesquiterpenes by Escherichia coli.
Biotechnol Bioeng 75:497-503.
Matsushika A, Inoue H, Kodaki T, Sawayama S. 2009. Ethanol production from xylose in engineered Saccharomyces
cerevisiae strains: current state and perspectives. Appl Microbiol Biotechnol 84: 37-53.
Maury J, Asadollahi MA, Møller K, Clark A, Nielsen J. 2005. Microbial isoprenoid production: an example of green
chemistry through metabolic engineering. Adv Biochem Eng Biotechnol 100:19-51.
McCaskill D, Croteau R. 1997. Prospects for bioengineering of isoprenoid biosynthesis. Adv Biochem Eng/Biotechnol
55:107-146.
McGravey DJ, Croteau R. 1995. Terpenoids metabolism. Plant cell. 7:1015-1026.
Misawa N. 2011. Pathway engineering for functional isoprenoids. Curr Opin Biotechnol 22:627-633.
Miyaoka H, Shida H, Yamada N, Mitome H, Yamada Y. 2002. Total synthesis of marine diterpenoid kalihine X.
Tetrahedron Lett 43:2227-2230Misra U. 2009. How Indian’s sandalwood oil trade got hijacked. Forbes India.
Morton. 2005. Life, reinvented. Wired Mag.
Mukaiyama T, Shiina I, Iwadare H, Saitoh M, Nishimura T, Ohkawa N, Sakoh H, N_ishimura K, Tani Y, Hasegawa M,
Yamada K, Saitho K. 1999. Asymmetric total synthesis of taxol. Chem Eur J 5:121-161.
Nautiyal OH. 2011. Analytical and Furrier transform infra red spectroscopy evaluation of sandalwood oil extracted
with various process techniques. J Nat Prod. 4:150-157.
Nevoigt E.2008. Progress in metabolic engineering of Saccharomyces cerevisiae. Microbiol Mol Biol Rev 72:379-412.
57
Nevoigt E, Kohnke J, Fisher CR, Alper H, Stahl U, Stephanpopulos G. 2006. Engineering of promoter replacement
cassettes for fine-tunung of gene expression in Sacchromyces cerevisae. Appl Environ Microbiol 72:5266-5273.
Newman JD, Marshall J, Chang M, Nowroozi F, Paradise E, Pitera D, Newman KL, Keasling JD. 2006. High-level
production of amorpha-4,11-diene in a two-phase partitioning bioreactor of metabolically engineered Escherichia coli.
Biotechnol Bioeng 95:684-691.
Nielsen J, Pronk JT. 2012. Metabolic engineering, synthetic biology and systems biology. FEMS Yeast Res 12:103.
Nielsen J, Keasling JD. 2011. Synergies between synthetic biology and metabolic engineering. Nat Biotechnol
29:693-695.
Nielsen J. 2009. Systems biology of lipid metabolism: from yeast to human. FEBS Lett 583:3905-3913.
Nielsen J, Jewett MC. 2007. Impact of systems biology on metabolic engineering of Saccharomyces cerevisiae.
FEMS Yeast Res 1-10
Nielsen J, Villadsen J, Liden G. 2003. Bioreaction Engineering Principles, second ed. Kluywer Plenum, New York.
Nielsen J. 2001. Metabolic engineering. Appl Microbiol Biotechnol.55:263-283.
Otero JM, Wongsangak W, Asadollahai M, Olivares R, Maury J, Farinelli L, Barlocher L, Osteras M, Schalk M, Clark
A, Nnilesen J. 2010. Whole genome sequencing of Saccharomyces cerevisiae: from genotype to phenotype for
improved metabolic engineering applications. BMC Genomics 11:723.
Otero JM, Panagiotou G, Olsson L. 2007. Fueling industrial biotechnology growth with bioethanol. Adv Biochem Eng
Biotechnol 108:1-40.
Olsson L, Nielsen J. 2000. The role of metabolic engineering in the improvement of Saccharomyces cerevisiae. Enz
Microb Technol. 26:785-792.
Ollson L, Larsen ME, Ronnow B, Mikkelsen JD, Nielsen J. 1997. Silencing MIG1 in Saccharomyces cerevisiae: effect
of the antisenseMIG1 expression and MIG1 gene disruption. Appl Environ Microbiol 63:2366-2371.
Palombi L. 2009. Beyond recombinant technology: synthetic biology and patentable subject matter. The Journal of
World Intellectual Propriety 12:371-401.
Paradise EM, Kirby J, Chan R, Keasling JD. 2008. Redirection of flux through the FPP branch-point in
Saccharomyces cerevisiae by down-regulating squalene synthase. Biotechnol Bioeng 100:371-378.
Partow S, Siewers V, Bjørn S, Nielsen J, Maury J, 2010. Characterization of different promoters for designing a new
expression vector in Saccharomyces cerevisiae. Yeast 27, 955-964.
Patel DJ, Suri AK, Jiang L, Fan P, Kumar RA, Nonin S. 1997. Structure, recognition and adaptive binding in RNA
adaptamer complexes. J Mol Biol 272:645-664.
Patil KR, Nielsen J. 2005. Uncovering transcriptional regulation of metabolism by using metabolic network topology.
Proc. Nat. Acad. Sci. 102:2685-2689
Patil KR, Akesson M, Nielsen J. 2004. Use of genome-scale microbial models for metabolic engineering. Curr Opin
Biotechnol 15:64:69.
Peralta-Yahya PP, Ouellet M, Chan R, Mukhopadhyay A, Keasling JD, Lee TS. Identification and microbial production
of a terpene-based advanced biofuel. Nat Comm 2:483.
Petranovic D, Vemuri G. 2009. Impact of yeast systems biology on industrial biotechnology. J Biotechnol 144:204211.
Pham HTB, Larsson G, Enfors SO. 1998. Growth and energy metabolism in aerobic fed-batch cultures of
Saccharomyces cerevisiae: simulation and model verification. Biotechnol Bioeng 60: 474-482.
Picaud S, Olofsson L, Brodelious M, Brodelious PE. 2005. Expression, purification and characterization of
recombinant amorpha-4,11-diene synthase from Artemisia annua L. Arch Biochem Biophys 436:215-226.
Pino JA, Marbot R, Fuentes V. 2006. Aromatics plant from Western Cuba IV. Composotion of the leaf oils of
Clausena lansium (Lour.) Seeklsand Swinglea glutinosa (Blanco) Merr. J Essent Oil Res 18:139-141
58
Polakowski T, Stahl U, Lang C. 1998. Overexpression of a cytosolic hydroxymethylglutaryl-CoA reductase leads to
squalene accumulation in yeast. Appl Microbiol Biotechnol 49:66-71.
Pronk JT, Steensma H, van Dijken JP. 1996. Pyruvate metabolism in Saccharomyces cerevisiae. Yeast 12:1607-33.
Renninger NS, Newman JD, Reiling KK. 2008. Fuel components, fuel compositions and methods of making and
using same. US2008/0092829 A1.
Ro DK, Ouellet M, Paradise EM, Burd H, Eng D, Paddon CJ, Newman JD, Keasling JD. 2008. Induction of multiple
pleiotropic drug resistance genes in yeast engineered to produce an increased level of anti-malarial drug precursor,
artemisinic acid. BMC Biotechnol 8:83.
Ro DK, Paradise EM, Ouellet M, Fisher KJ, Newman KL, Ndungu JM, Ho KA, Eachus RA, Ham TS, Kirby J, Chang
MCY, Withers ST, Shiba Y, Sarpong R, Keasling JD. 2006. Production of the anitmalarial drug precursor artemisinic
acid in engineered yeast. Nature 440:940-943.
Ritter SK. 2008. Plant Biochemistry. Lignocellulose: a complex biomaterial. Chemical Engineering News 86:15
Rude MA, Schirmer A. 2009. New microbial fuels: a biotech perspective. Curr Opin Microbiol. 12: 274-281.
Russo S, Berkovitz Siman-Tov, Poli G. 1995. Yeast: from genetics to biotechnology. J Environ Pathol Toxicol Oncol.
14:133-157.
Sauer .2001. Evolutionary engineering of industrially important microbial phenotypes. Adv Biochem Eng Biotechnol
73:129-169.
Scallen TJ, Sanghvi A. 1983. Regulation of three key enzymes
phosphorylation/dephosphorylation. Proc Natl Acad Sci USA 80:2477-2480.
in
cholesterol
metabolism
by
Schalk M. 2011. Method for producing Alpha-santalene. US Pat 2011/008836 A1.
Siddiqui MS, Thodey K, Trechard I, Smolke CD. 2012. Advancing secondary metabolite biosynthesis in yeast with
synthetic biology. FEMS Yeast Res 12: 144-170.
Sims B. BASF leads investment round in specialty chemical firm Allylix. 2012. Biorefining magazine.
Sikkema J, de Bont JAM, Poolman B. 1995. Mechanisms of membrane toxicity of hydrocarbons. Microbiol Rev
59:201-222.
Sikkema J, de Bont JAM, Poolman B. 1994. Interaction of cyclic hydrocarbons with biological membranes. J Biol
Chem 18:8022-8028.
Siewers V, Mortensen U, Nielsen J. 2010. Genetic enrineering tools for Saccharomyces cerevisiae . Manual of
industrial microbiology and biotechnology. 3 ed. ASM press, Washington DC.
rd
Smolke CD, Tyo KE. Systems biology: emerging methodologies to catalyze the metabolic engineering design cycle.
Metab Eng 14:187-188.
Solem C, Jensen R. 2002. Modulation of gene expression made easy. Appl Environ Microbiol 68:2397-2403.
Steen EJ, Kang Y, Bokinsky G, Hu Z, Schimer A, Mcclure A, Cardayre SB, Keasling JD. 2010. Microbial production
of fatty-acid-derived fuels and chemicals from plant biomass. Nature 463:559-562.
Stephanopoulos G. 2007. Challenges in engineering microbes for biofuels production. Science 315: 801-804.
Stepanopoulos G. 1999. Metabolic fluxes and metabolic engineering. Metab Eng 1:1-10.
Stephanopoulos GN, Aristidou AA, Nilesen J. 1998. Metabolic engineering principles and methodologies. Academic
press, USA.
Stephanopoulos G, Vallino JJ. 1991. Network rigidity and metabolic engineering in metabolite overproduction.
Science 252:1675-1681.
Steensma HY, Verrips CT, Vindelov J, Pronk JT. 2000. An interlaboratory comparison of physiological and genetic
properties of four Saccharomyces cerevisiae strains. Enzyme Microb Technol 26: 706-714.
59
Sun J, Shao Z, Zhao H, Nair N, Wen F, Xu JH, Zhao H. 2012. Cloning and characterization of a panel of constitutive
promoters for applications in pathway engineering in saccharomyces cerevisiae. Biotechnol Bioeng in press
Tait J, Castle D. 2012. Balanced regulation of synthetic biology. Nature 484: 37.
Takahashi S, Yeo Y, Greenhagen BT, McMullin T, Song L, Maurina-Brunker J, Rosson R, Noel JP, Chappell J. 2007.
Metabolic engineering of sesquiterpene metabolism in yeast. Biotechnol Bioeng 97:170-181.
Teisserire PJ. 1994. Chemistry of fragrant substances. VCH Publ., New York.
The Economist. 2010. Chemistry goes green, behind scenes, industrial biotechnology is getting going at last. The
Economist Newspaper Limited, London Jul 1, 2010.
van Dijken J P, Bauer J, Brambilla L, Duboc P, Francois JM, Gancedo C, Giuseppin ML, Heijnen JJ, Hoare M, Lange
HC, Madden EA, Niederberger P, Nielsen J, Parrou JL, Petit T, Porro D, Reuss MN, van Riel N, Rizzi M,
Van Gulik WM, Heijnen JJ. 1995. A network stoichiometry analysis of microbial growth and product formation.
Biotechnol and Bioeng 48:681-698.
Van Vleet JH, Jeffries TW. 2009. Yeast metabolic engineering for hemicellulosic ethanol production. Curr Opin
Biotech 20: 300-306.
van Maris AJ, Winkler AA, Kuyper M, de Laat WT, van Dijken JP, Pronk JT. 2007. Development of efficient xylose
fermentation in Saccharomyces cerevisiae: xylose isomerase as a key component. Adv Biochem Eng Biotechnol 108:
179-204.
Venter C, Choen D. 2004. The century of Biology. New Perspective Quarterly 21:73-77
Verwaal R, Jiang Y, Daran JM, Sandmann G, van den Berg J, van Ooyen AJJ. 2010. Heterologous carotenoid
production in Saccharomyces cerevisiae induces the pleiotropic drug resistance stress response. Yeast 27:985-988.
Vik Å, Rine J. 2001. Upc2p and Ecm22p, dual regulators of sterol biosynthesis in Saccharomyces cerevisiae. Mol
Cell Biol 21: 6395-6405.
Wang C, Kim JY, Choi ES, Kim SW. 2011a Microbial production of farnesol (FOH): current states and beyond.
Process Biochem 46:1221-1229.
Wang C, Yoon H.S, Jang JH, Chung RY, Kim YJ, Choi SE, Kim WS. 2011b. Metabolic engineering of Escherichia coli
for α-farnesene production. Metab Eng 13: 648-655.
Watanabe SA, Abu Saleh SP, Pack N, Annaluru T, Kodaki T, Makino K. 2007. Ethanoio production from xylose by
recombinant Saccharomyces cerevisiae expressing protein engineered NADH-preferring xylose reductase from
Pichia stipitis. Microbiology 153:3044-3054.
Westfall P, Pitera DJ, Lenihan JR, Eng D, Woolard FX, Regentin R, Horning T, Tsuruta H, Melis DJ, Owens A, Fickers
S, Diola D, Benjamin KR, Keasling JD, Leavell MD, McPhee DJ Renninger NS, Newman JD, Paddon CJ. 2011.
Production of amorphadiene in yeast, and its conversion to dihydroartemisinic acid, precursor to the antimalarial
agent artemisinin. Proc Natl Acad Sci 109:111-118.
Wetterstrand KA. 2012. DNA sequencing costs: data from the NHGRI large-scale genome sequencing program.
Available at: http://www.genome.gov/sequencingcosts/.
Whited GM, Feher FJ, Benko DA, Cevin MA, Chotani GK, McAuliffe JC, LaDuca RJ, Ben-Shoshan EA, Stanford KJ.
2010. Development of a gas-phase bioprocess for isoprene-monomer production using metabolic pathway
engineering. Industrial Biotechnol 6:152-163.
You L, Cox III RS, Weiss R, Arnold FH. 2004. Programmed population control by cell-cell communication and
regulation. Nature 428:868-871.
Young E, Alper H. 2010. Synthetic biology: tools to design, build, and optimize cellular processes. J Biomed
Biotechnol Article ID 130781, 12 pages
Yoshikuni Y, Ferrin TE, Keasling JD. 2006. Design divergent evolution of enzyme function. Nature 440:1078-1082.
Zhang F, Keasling J. 2011. Microbes and Metabolism. Biosensors and their application in microbial metabolic
engineering. Cell 19:323-329.
60
Zhang F, Rodriguez S, Keasling JD.2011. Metabolic engineering of microbial pathway for advanced biofuel
production. Curr Opin Biotechnol 22:775-783.
Zamore PD, Haley B. 2005. Ribo-gnome: the big world of small RNAs. Science 309:1519-1524.
Zhao J, Nan P, Zhong Y. Chemical composition of the essential oils of Clausena lansium from Hainan Island, China. Z
Naturforsch 59c:153-156.
61
PAPER I
Dynamic control of gene expression in Saccharomyces
cerevisiae engineered for the production of plant sesquiterpene
α-santalene in fed batch mode.
Scalcinati G, Knuf C, Partow P, Chen Y, Maury J, Schalk M, Daviet L, Nielsen J,
Siewers V.
Metabolic Engineering. 2012. 14 (2): 91-103.
62
Metabolic Engineering ] (]]]]) ]]]–]]]
Contents lists available at SciVerse ScienceDirect
Metabolic Engineering
journal homepage: www.elsevier.com/locate/ymben
Dynamic control of gene expression in Saccharomyces cerevisiae engineered
for the production of plant sesquitepene a-santalene in a fed-batch mode
Gionata Scalcinati a,1, Christoph Knuf a,1, Siavash Partow a, Yun Chen a, Jérôme Maury b, Michel Schalk c,
Laurent Daviet c, Jens Nielsen a,n, Verena Siewers a
a
Department of Chemical and Biological Engineering, Chalmers University of Technology, SE-412 96 Göteborg, Sweden
Center for Microbial Biotechnology, Department of Systems Biology, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark
c
Firmenich SA, Corporate R&D Division, CH-1211 Geneva 8, Switzerland
b
a r t i c l e i n f o
abstract
Article history:
Received 2 September 2011
Received in revised form
11 January 2012
Accepted 26 January 2012
Microbial cells engineered for efficient production of plant sesquiterpenes may allow for sustainable
and scalable production of these compounds that can be used as e.g. perfumes and pharmaceuticals.
Here, for the first time a Saccharomyces cerevisiae strain capable of producing high levels of a-santalene,
the precursor of a commercially interesting compound, was constructed through a rationally designed
metabolic engineering approach. Optimal sesquiterpene production was obtained by modulating the
expression of one of the key metabolic steps of the mevalonate (MVA) pathway, squalene synthase
(Erg9). To couple ERG9 expression to glucose concentration its promoter was replaced by the HXT1
promoter. In a second approach, the HXT2 promoter was used to express an ERG9 antisense construct.
Using the HXT1 promoter to control ERG9 expression, it was possible to divert the carbon flux from
sterol synthesis towards a-santalene improving the productivity by 3.4 fold. Combining this approach
together with the overexpression of a truncated form of 3-hydroxyl-3-methyl-glutaryl-CoA reductase
(HMGR) and deletion of lipid phosphate phosphatase encoded by LPP1 led to a strain with a
productivity of 0.18 mg/gDCW h. The titer was further increased by deleting DPP1 encoding a second
FPP consuming pyrophosphate phosphatase yielding a final productivity and titer, respectively, of
0.21 mg/gDCW h and 92 mg/l of a-santalene.
& 2012 Published by Elsevier Inc.
Keywords:
Metabolic engineering
Isoprenoids
Farnesyl diphosphate
Ergosterol
Squalene synthase
Saccharomyces cerevisiae
1. Introduction
The production of plant sesquiterpenes by microbial fermentation is an environmentally friendly and attractive alternative to
the commonly used chemical synthesis and plant extraction
(Chang and Keasling., 2006; Rohlin et al., 2001; Ajikumar et al.,
2008; Kirby and Keasling, 2009, Wang et al., 2011). Sesquiterpenes are mainly used by the chemical industry for production of
fragrances and aroma chemicals (Daviet and Schalk, 2010). They
represent a diverse class of secondary metabolites, the precursors
of which are natively present in the metabolic network of
Saccharomyces cerevisiae (Förster et al., 2003). Sesquiterpenes
are naturally produced in S. cerevisiae through the mevalonate
(MVA) pathway from multiple condensation of isopentenyl
diphosphate (IPP) and dimethyllallyl diphosphate (DMAPP) leading to farnesyl diphosphate (FPP), the universal precursor unit of
all sesquiterpenes (C15) (Maury et al., 2005; Withers and Keasling,
n
Corresponding author. Fax: þ46 31 772 38 01.
E-mail address: [email protected] (J. Nielsen).
1
Contributed equally to this research.
2007). Nowadays, most fragrance compounds are produced via
plant and microbe extraction (Howes et al., 2004), and by total
chemical synthesis or semi-synthesis (Corey et al., 1957; Julia,
1976; Janssens et al., 1992). Traditional methods of extraction are
limited by low yields and high costs. Here, microbial biosynthesis
offers several advantages: it (i) avoids formation of racemic
mixtures providing enantiomerically pure products through enzymatic biocatalysis, (ii) reduces process costs using inexpensive
sugar based carbon sources, (iii) increases sustainability avoiding
harvesting and extraction from natural sources and thus reducing
environmental footprint, (iv) increases yield and productivities
using genetic manipulation of the heterologous hosts and (v) is
compatible with scalable high density fermentation processes.
a-Santalene is the precursor of a-santalol, one of the main
components of East Indian sandalwood oil (Corey et al., 1957;
Baldovini et al., 2010). Sandalwood oil is commonly used in
cosmetic, perfumery and aromatherapy industries and has
recently been identified as a potential chemotherapeutic and
chemopreventive agent against skin cancer (Dwivedi et al.,
2003). a-santalene is produced enzymatically in a one-stepconversion from farnesol diphosphate catalyzed by a plant santalene synthase (Schalk, 2011). Here, an efficient S. cerevisiae strain
1096-7176/$ - see front matter & 2012 Published by Elsevier Inc.
doi:10.1016/j.ymben.2012.01.007
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
2
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
Nomenclature
FPP
FOH
SQS
HMGR
MVA
SanSyn
Log P
Farnesyl diphosphate
(E,E)-farnesol
squalene synthase
capable of reaching relevant titers and productivities of a-santalene during an optimized fermentation process is constructed.
First, yeast was engineered to produce a-santalene by introducing
a heterologous santalene synthase gene (SanSyn) derived from
Clausena lansium (wampee) that catalyzes the conversion of FPP
to a-santalene (Schalk, 2011). Metabolic engineering has been
extensively applied to manipulate metabolic fluxes and enhance
the microbial production of sesquiterpene compounds (Ro et al.,
2006; Shiba et al., 2007; Takahashi et al., 2007; Asadollahi et al.,
2008, 2009, 2010; Kirby et al., 2008; Ma et al., 2011). In order to
increase the precursor pool for isoprenoid synthesis enabling
efficient conversion to the target compound a-santalene, two of
the main regulatory steps of the MVA pathway catalyzed by
3-hydroxyl-3-methyl-glutaryl-CoA reductase (HMGR) and squalene synthase (SQS) were optimized by introducing genetic
modifications that enable to channel the flux towards a-santalene
synthesis. The conversion of 3-hydroxyl-3-methyl-glutaryl-CoA
into mevalonate catalyzed by HMGR is one of the most studied
key regulatory steps in the MVA pathway and is considered the
main flux controlling step (Scallen and Sanghvi, 1983; Basson
et al., 1986). In yeast, two isoforms of HMGR exist encoded by
genes HMG1 and HMG2 (Basson et al., 1986) and their activity is
subject to tight regulation including feedback regulation and
cross-regulation (Hampton and Rine, 1994; Hampton et al.,
1996; Brown and Goldstein, 1980). HMGR is composed of an
interspecies conserved catalytic domain and a variable membrane
anchoring region also referred to as sterol sensing domain (SSD).
Hmg1 regulation acts at the level of the SSD domain through a
complex mechanism leading to protein degradation (Nielsen,
2009). Over-expression of a truncated form of Hmg1 lacking the
SSD domain bypasses this post-transcriptional control and results
in a constitutively active non-membrane bound form (Donald
et al., 1997; Polakowski et al., 1998). This strategy has been
3-hydroxy-3-metyl-glutaryl-coenzyme A reductase
mevalonate
santalene synthase gene
logarithm (base 10) of partition coefficient
extensively used to increase the flow through the MVA pathway in
order to produce isoprenoid derived compounds (Jackson et al., 2003;
Ro et al., 2006; Kirby et al., 2008; Asadollahi et al., 2009, 2010; Farhi
et al., 2011). The second key step in the MVA pathway is represented
by SQS, as this controls the flux of FPP towards sterols or non-sterol
sesquiterpenes. FPP is a pivotal intermediate as it is a common
precursor for production of essential compounds such as dolichol,
ubiquinone, isoprenylated proteins and ergosterol (Daum et al., 1998)
and its intracellular concentration is tightly regulated at different
levels (Goldstein and Brown, 1990). During normal growth conditions, most of the FPP is converted into ergosterol due to the fact that
the cellular demand for sterols is greater compared to the demand for
non-sterol FPP derived compounds (Kennedy et al., 1999). Strategies
to increase sesquiterpene production based on the disruption of the
main FPP consuming reaction catalyzed by squalene synthase produced lethal mutants and restoration of viability requires ergosterol
supplementation resulting in an economically not feasible process for
industrial purposes (Takahashi et al., 2007). Here, we undertook a
genetic engineering approach to balance SQS activity during a
fermentation process. Previous attempts to control SQS (encoded by
ERG9) expression diverting the flow from sterol components to
desired FPP-derived compounds were mainly based on replacement
of the native ERG9 promoter with the methionine-repressible MET3
promoter (Ro et al., 2006; Paradise et al., 2008; Asadollahi et al., 2008,
2009, 2010). However, industrial scale development of this system is
limited by the cost of the repressor and its possible consumption by
the cell. Instead, we aimed at coupling ERG9 expression to the glucose
concentration in the media.
The production capacity of the engineered strains is evaluated
through a fermentation process coupling biochemical production
to biomass formation that allows capturing the water insoluble
compound during production resulting in an efficient S. cerevisiae
cell factory for biosynthesis of sesquiterpenoid fragrances.
Table 1
S. cerevisiae strains used in this study.
Strain
Genotype
Plasmid
c
CEN.PK113-5D
MATa MAL2-8 SUC2 ura3-52
YIP-M0-04
MATa MAL2-8c SUC2 ura3-52 PERG9D::kanMX-PMET3
SCICK00
SCICK01
SCICK03
SCICK05
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP PERG9D::loxP-PMET3
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP PERG9D::loxP-PHXT1
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP PERG9D::loxP-PERG9
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP PERG9D::loxP-PERG9
YMRWdelta15D::PHXT2-ERG9as-TCYC1-loxP
MATa MAL2-8c SUC2 ura3-52::pSF011-PHXT1
MATa MAL2-8c SUC2 ura3-52::pSF011-PTEF1M2
MATa MAL2-8c SUC2 ura3-52::pSF011-PHXT2
MATa MAL2-8c SUC2 ura3-52::pSF011-PMET3
MATa MAL2-8c SUC2 ura3-52::pSF011-PERG9
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP PERG9D::loxP-PHXT1
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP PERG9D::loxP-PERG9
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP PERG9D::loxP-PERG9
YMRWdelta15D::PHXT2- asERG9-TCYC1-loxP
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP PERG9D::loxP-PMET3
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP dpp1D::loxP PERG9D::loxP-PHXT1
MATa MAL2-8c SUC2 ura3-52 lpp1D::loxP dpp1D::loxP PERG9D::loxP-PHXT1
SCICK06
SCICK08
SCICK09
SCICK10
SCICK11
SCICK12
SCICK13
SCICK14
SCICK15
SCICK16
SCICK17
Reference
P. Kötter, University
of Frankfurt,
Germany
Asadollahi et al.,
2008
This study
This study
This study
This study
pICK01
pICK01
pICK01
pICK01
pICK01
This
This
This
This
This
This
This
This
study
study
study
study
study
study
study
study
This study
This study
This study
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
3
2. Materials and methods
2.2. Plasmid construction
2.1. Strains and maintenance
To construct integrative plasmids carrying the lacZ gene under
control of different promoters plasmid pSF011 (Partow et al.,
2010) was used. Promoters PERG9, PMET3, PHXT1, and PHXT2, and were
PCR amplified from genomic DNA of S. cerevisiae CEN.PK113-5D,
and PTEF1M2 was amplified from plasmid p416TEF1M2 (kindly
provided by G. Stephanopoulos, Massachusetts Institute of Technology, Cambridge, MA, USA) using primers 1 to 6 and 50 to 53
(Table 2), restricted with NotI/HindIII and ligated into NotI/HindIII
cut pSF011 resulting in formation of plasmids pSF011-PERG9,
pSF011-PMET3 and pSF011-PHXT1, pSF011-PHXT2 and pSF011PTEF1M2, respectively.
To construct the a-santalene expression vector the SanSyn
gene was amplified by PCR from plasmid Cont2B-27-pET101
Strains used in this study are listed in Table 1. Strains were
maintained on YPD plates containing 10 g/l yeast extract, 20 g/l
casein peptone, 20 g/l glucose and 20 g/l agar. Plasmid carrying
strains were selected on synthetic dextrose (SD) agar containing
6.9 g/l yeast nitrogen base w/o amino acids (Formedium, Hunstanton, UK), 0.77 g/l complete supplement mixture (CSM) w/o
uracil (MP Biomedicals, Solon, OH, USA), 20 g/l glucose, and 20 g/l
agar and counter-selected on SD plates supplemented with
30 mg/l uracil and 750 mg/l 5-fluoroorotic acid (Formedium).
Strains containing the kanMX cassette were selected on YPD
plates containing 200 mg/l G418 (Formedium).
Table 2
Oligonucleotide primers used in this studya.
No.
Name
Sequence (50 -30 )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
HindIII_ERG9_f
NotI_ERG9_r
HindIII_MET3_f
NotI_MET3_r
HindIII_HXT1_f
NotI_HXT1_r
santa_f
santa_r
tHMG1_up
tHMG1_down
Int_check_f
Int_check_r
LacZ_r
LPP_up_f
LPP_up_r_tail
LPP_dw_f_tail
LPP_dw_r
Pr-b-Kan0
Kan30 int
Kan50 int
dKan30
LPP1 verif_up
LPP1 verif_dw
loxP_ERG9_f
ERG9d_r
loxP_HXT1_f
ERG9_Hxt1_r
ERG9_loxP_f
loxP_r
Erg9_fr_f
Erg9_fr_r
DPP-1-fw
DPP-1-rev
DPP-2-fw
DPP-2-rev
KanMx-1-fw
KanMx-1-rev
KanMx-2-fw
KanMx-2-rev
Amp_fw
Amp_rv
14up_HXT2_f
ERG9as_HXT2_r
ERG9as_f
CYC1t_ERG9as_r
ERG9as_CYC1t_f
loxP_CYC1t_r
loxP_f
14down_loxP_r
HindIII_HXT2_f
NotI_Hxt2_r
HindIIITEF1_M2_f
NotI_TEF1_M2_r
CAACAAAAGCTTCCCATCTTCAACAACAATACC
CAACAAGGCGGCCGCTGTGTGTGTGTGATATGTGACGT
CAACAAAAGCTTGTATAAGGTGAGGGGGTCCACAG
CAACAAGGCGGCCGCGTTAATTATACTTTATTCTTGTTATTATTATACtttc
CAACAAAAGCTTTGCAGGTCTCATCTGGAATATAATTCC
CAACAAGCGGCCGCGATTTTACGTATATCAACTAGTTGACGATTATqG
GTTGTTGCGGCCGCAAAACAATGTCAACTCAACAAGTTTCATCAG
GTTGTTTTAATTAACTAATCGTCAAGCTTAACGGG
GTTGTTGGATCCAAAACAATGGCTGCAGACCAATTGGTG
GTTGTTGCTAGCTTAGGATTTAATGCAGGTGACG
GGTTCCGCGCACATTTCCCC
GGAACTCTTGTTGTTCTTTGGAG
GGGATCTGCCATTGTCAGAC
AAGGATGATCTCTGTCATGG
GATCCCCGGGAATTGCCATGTGTTAGGGCAGCATTTATGC
GCAGGGATGCGGCCGCTGACGCACTCCAAGCGGACATTCAAG
GAAGTATGTCTCTTTTCGCC
CATGGCAATTCCCGGGGATCCCCTTAATATAACTTCGTATAATGTATGC
CCATGAGTGACGACTGAATCCGG
GCAAAGGTAGCGTTGCCAATG
GTCAGCGGCCGCATCCCTGCCGACTCACTATAGGGAGACCG
TAGTTGCCACGTGAAACCTGACAAC
AATTTCATCGGTATTTTGGCTTCGG
CGAAGTTATTAGGTGATATCAGATCCACTGCCCATCTTCAACAACAATACCG
GTCGTAGTCGTGGACGGTTTGC
CGAAGTTATTAGGTGATATCAGATCCACTTGCAGGTCTCATCTGGAATATAATTCC
GCTGCCTTCATCTCGACCGGATGCAATGCCAATTGTAATAGCTTTCCCATGATTTTACGTATATCAACTAGTTGACGATTATG
GAGTGAACCTGCTGCCTGGCGTGCTCTGACTCAGTACATTTCATAGCCCAGTACGCTGCAGGTCGACAACC
AGTGGATCTGATATCACCTAATAACTTCG
CCTTGCTTACACAGAGTGAACCTGCTGCCTGGC
CTTCAGCTTCAAAGCTGCCTTCATCTCGACCG
AGGGCACGTTATCAATTGT
CAGCGTACGAAGCTTCAGAGAAACTCGTACTGAACCAAG
GTGATATCAGATCCACTAGTACAGACATCATTTCGTCGA
AACTTCTAAGGCTTTCGTGT
CTGAAGCTTCGTACGCTG
TCACCATGAGTGACGACTGA
TTCCAACATGGATGCTGAT
CTAGTGGATCTGATATCAC
GTGGGTTACATCGAACTGGATC
CATCCATAGTTGCCTGACTGC
GATTTGATGACTGTTTCTCAAAACTTTATGTCATTTTCTTACACCGCATATGATTCTACCGATGTAATACAAAAATG
TTATTGAATTCCACAAATTGAAACTATGTTGCTTTATAAGTCTTTTTGTAA
GTTTCAATTTGTGGAATTCAATAA
TTCCTTTTCGGTTAGAGCGGATCTGGAAATTAGGACAGGGGC
GCCCCTGTCCTAATTTCCAGATCCGCTCTAACCGAAAAGGAA
GGTTGTCGACCTGCAGCGTACCTTCGAGCGTCCCAAAACCTT
GTACGCTGCAGGTCGACAACC
GATAACCGCGAAGATTTATAATGGTTTATCGGTTGCATTTTCCATGAGTAAGTGGATCTGATATCACCTAATAACTTCG
CAACAAAAGCTTTTCTACCGATGTAATACAAAAATG
CAACAAGGCGGCCGCTATGTTGCTTTATAAGTCTTTTTGTAA
CAACAAAAGCTTGCACACACCACGGCTCTAAAG
CAACAAGCGGCCGCTTTTCTAGAAAACTTGGATTAGATTGC
a
Restriction enzyme recognition sites are underlined.
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
4
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
(Schalk, 2011), using primers 7 and 8, cut with NotI/PacI and
ligated into NotI/PacI restricted vector pSP-G1 (Partow et al.,
2010). Subsequently, tHMG1 was PCR amplified using genomic
DNA of S. cerevisiae CEN.PK113-5D as template and primers 9 and
10, cut with BamHI/NheI and ligated into the same vector after
restriction with the respective enzymes. This resulted in formation of expression plasmid pICK01.
2.3. Strain construction
Strains carrying a genomic integration of lacZ under control of
different promoters were constructed by transforming
CEN.PK113-5D with the NcoI restricted integrative plasmids
pSF011-PERG9, pSF011-PMET3, pSF011-PHXT1, pSF011-PHXT2 and
pSF011-PTEF1M2, respectively, resulting in formation of strains
SCICK11, SCICK10, SCICK06, SCICK09 and SCICK08, respectively.
Correct integration at the ura3-52 locus was verified using
primers 11 and 12. To test for tandem integration of the plasmid,
primers 11 and 13 were used. To exclude additional integrations
of the plasmid, strains were subjected to Southern blot analysis.
The loxP flanked kanMX cassette in strain YIP-M0-04 was
excised with help of the Cre recombinase expression plasmid
pSH47 as described by Güldener et al. (1996). To delete LPP1 in
this strain a bipartite gene targeting strategy was applied (Erdeniz
et al., 1997). The 50 and 30 region of the gene were amplified by
PCR using primer pairs 14/15 (fragment 1) and 16/17 (fragment
2), respectively, and genomic DNA of CEN.PK113-5D as template.
The 50 and the 30 part of the kanMX cassette were amplified from
plasmid pUG6 (Güldener et al., 1996) using primer pairs 18/19
(fragment 3) and 20/21 (fragment 4), respectively. Complementary primer tails allowed for the combination of fragments 1 and
3 by fusion PCR. Likewise, fragments 2 and 4 were fused to each
other. Cells were transformed with both fusion PCR fragments
and integration of the kanMX cassette at the LPP1 locus was tested
by PCR using primers 22 and 23. Subsequent excision of the
kanMX cassette led to formation of strain SCICK00.
To replace the ERG9 controlling MET3 promoter in SCICK00 the
ERG9 promoter and the HXT1 promoter were amplified from
genomic DNA by PCR using primer pairs 24/25 and 26/27,
respectively. In addition, the kanMX cassette was amplified in a
PCR containing primers 28 and 29. The marker cassette was
combined with either of the two promoters by fusion PCR and
the resulting fragments were amplified once more using primers
30 and 31 in order to extend the flanking regions for genomic
integration. Transformation of SCICK00 with these fragments and
subsequent excision of the kanMX cassette resulted in strains
SCICK01 and SCICK03, respectively.
For genomic integration of an ERG9 antisense expression
cassette, four PCR fragments were generated containing the
HXT2 promoter (primers 42/43, fragment 5), a fragment of ERG9
(primers 44/45, fragment 6), the CYC1 terminator (primers 46/47,
fragment 7) and the kanMX cassette flanked by loxP sites (primers
48/49, fragment 8). The ERG9 fragment comprised the first 412 bp
of the coding sequence and 99 bp upstream of the start codon.
Fragments 5, 6 and 7 and fragment 7 and 8 were combined by
fusion PCR and used to transform SCICK03. 50 extensions of
primers 42 and 49 allowed for integration by homologous
recombination at YMRWdelta15 (Flagfeldt et al., 2009). Excision
of the kanMX cassette resulted in strain SCICK05.
For deletion of DPP1 in strain SCICK01, bipartite gene targeting
was applied as described above. Here, primer pairs 32/33 and 34/
35 were used to amplify the 50 and 30 region of DPP1 and primer
pairs 36/37 and 38/39 were employed for amplification of the 50
and 30 part of the kanMX cassette. Integration and following
excision of the resistance marker led to formation of strain
SCICK16.
By transforming SCICK00, SCICK01, SCICK03, SCICK05 and
SCICK16 with plasmid pICK01 strains SCICK15, SCICK12, SCICK13,
SCICK14 and SCICK17 were constructed.
2.4. Southern blot analysis
Genomic DNA was cut with HindIII, separated on a 1% agarose
gel and transferred to a HybondTM-N þ membrane (GE Healthcare, Uppsala, Sweden) according to the supplier’s instructions. As
probe, a fragment of the ampicillin resistance gene in vector
pSF011 was amplified by PCR using primers 40 and 41. The
AlkPhos Direct Labeling and Detection system (GE Healthcare)
was applied using CDP-Star as detection reagent and a ChemiDoc
XRS system (Bio-Rad Laboratories, Hercules, CA, USA) for chemiluminescence imaging.
2.5. Media and growth conditions
For batch cultivations, a previously described (Verduyn et al.,
1992) mineral salts medium was used consisting of the following
(per liter): (NH4)2SO4, 5 g; KH2PO4, 3 g; MgSO4 7H2O, 0.50 g;
Antifoam 289 (A-5551, Sigma-Aldrich, St. Louis, MO, USA),
0.05 ml; trace metals, 1 ml and vitamins, 1 ml. The trace metal
solution consisted of the following (per liter): EDTA (sodium salt),
15.0 g; ZnSO4 7H2O, 0.45 g; MnCl2 2H2O, 1 g; CoCl2 6H2O, 0.3 g;
CuSO4 5H2O, 0.3 g; Na2MoO4 2H2O, 0.4 g; CaCl2 2H2O, 0.45 g;
FeSO4 7H2O, 0.3 g; H3BO3, 0.1 g and KI, 0.1 g. The pH of the trace
metal solution was adjusted to 4.0 with 2 M NaOH prior to heat
sterilization. The vitamin solution contained (per liter): biotin,
0.05 g; p-amino benzoic acid, 0.2 g; nicotinic acid, 1 g; Ca-pantothenate, 1 g; pyridoxine-HCl, 1 g; thiamine-HCl, 1 g and myoinositol, 25 g. The pH of the vitamin solution was adjusted to
6.5 with 2 M NaOH. The vitamin solution was filter sterilized and
stored at 4 1C. This medium was supplemented with 30 g/l
glucose. The medium used for shake flask cultivation had the
same composition as described above, but the (NH4)2SO4 concentration was increased to 7.5 g/l, and the KH2PO4 concentration
to 14.4 g/l. The glucose concentration was 20 g/l and the pH was
adjusted to 6.5 prior autoclaving. The feed composition used for
fed-batch cultivation had the same composition as described
above, but the (NH4)2SO4; KH2PO4; MgSO4 7H2O, vitamin solution, and trace metal solution concentrations were increased 10
times; the glucose concentration was 200 g/l.
2.6. Inoculum preparation
The seed cultures for the cultivations were grown at 30 1C in
500-ml shake flasks containing 100 ml of culture with agitation in
an orbital shaker at 100 rpm. Pre-cultures were used to inoculate
the fermentors to a final dry weight of 1 mg/l. All cultivations
were performed in duplicate.
2.7. Shake flask cultivation
Cultivations were carried out in 500 ml baffled Erlenmeyer
flasks with four diametrically opposite baffles and side necks for
aseptic sampling. The flasks were prepared with 100 ml medium
as described above. Cultures were incubated with agitation in an
orbital shaker at 100 rpm and the temperature was controlled at
30 1C.
2.8. Fed-batch mode in shake flasks
Shake flasks in fed-batch mode were realized using the
FeedBeads polymer-based slow-release technique as previously
described (Jeude et al., 2006). Media were prepared as described
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
5
2.11. b-Galactosidase activity assay
above without initial glucose content. Four sterile silicone elastomer disks Ø12 mm (Kühner AG, Basel, Switzerland) containing
glucose crystals were added immediately before inoculation to
250 ml Erlenmeyer flasks containing 25 ml of medium.
The enzyme activity assay was performed as described earlier
(Flagfeldt et al., 2009).
2.9. Fed-batch operation
2.12. Analysis of sesquiterpenes
The aerobic fed-batch process was performed in 2.5 l Applikon
vessels (Applikon, Schiedam, The Netherlands) with a working
volume of 1.0 l. Agitation at 600 rpm was maintained using an
integrated stirrer (DasGip, Jülich, Germany) and the temperature
kept at 30 1C. The rate of aeration was set to 0.6 l/min. The pH of
the medium was maintained at 5.0 by automatic addition of 2 N
KOH during the batch phase and 10% NH4OH during the feed
phase. The temperature, agitation, gassing, pH and composition of
the off-gas were monitored and controlled using the DasGip
monitoring and control system. Dissolved oxygen concentration
was monitored with an autoclavable polarographic oxygen electrode (Mettler Toledo, Columbus, OH, USA) and kept above 30%
via stirrer speed and gas flow rate using the DasGip control
system. The effluent gas from the fermentation was analyzed for
real-time determination of oxygen and CO2 concentration by
DasGip fedbatch pros gas analysis systems with the off gas
analyzer GA4 based on zirconium dioxide and two-beam infrared
sensor.
The fed-batch cultures were initiated as batch cultures using
30 g/l glucose. Feeding with fresh medium commenced only after
residual ethanol produced from the glucose consumption phase
was completely depleted. A feed strategy was designed keeping
the volumetric growth rate constant (Nielsen et al., 2003). An
exponential feed rate n(t) (l/h) was calculated according to:
Sequiterpene production during the course of fermentation
was determined as described previously (Asadollahi et al., 2010).
Samples from the organic layer were centrifuged 5 min at 5000 g
and the supernatants were analyzed by gas chromatographymass spectrometry (GC/MS) with a DSQ II single quadrupole mass
spectrometer (Thermo Scientific, Waltham, MA). Analytes from
1 mL sample were separated on a SLB-5 ms capillary column
(15 m, 0.25 mm i.d., 0.25 mm film thickness; Supelco, Bellefonte,
PA, USA) using helium as carrier gas at a flow rate of
1.2 ml min 1. A split/splitless injector was used in the splitless
mode. The initial oven temperature was 80 1C and the injector
temperature was 250 1C. The oven temperature was increased to
120 1C at a rate of 10 1C/min and subsequently increased to 160 1C
at a rate of 3 1C/min. The oven temperature was finally increased
to 270 1C at a rate of 10 1C/min and held for 5 min at this
temperature. Full mass spectra were generated by scanning m/z
range within 40–500 for metabolite identification, a-santalene
and E,E-farnesol were identified comparing mass spectra and
retention time with the available authentic standards; trans-abergamotene by comparison with library spectra using NIST mass
spectra search program (http://chemdata.nist.gov/index.html).
Quantification of a-santalene and E,E-farnesol was carried out
using standard curves, trans-a-bergamotene was quantified with
a correction factor determined using a-humulene as internal
standard.
nðtÞ ¼
Y xs m0
x0 V 0 expðm0 tÞ
sf s0
where x0, s0 and V0 were the biomass density (gDCW/l), the
substrate concentration (g/l) and the reactor volume (l) at the
start of the feed phase, Yxs was the respiratory yield coefficient
(g glucose/gDCW); sf was the concentration of the growth limiting
substrate (g glucose/l) in the reservoir; m0 the was the specific
growth rate (h 1) during the feed phase and t the feeding time.
According to the equation above the feed was increased exponentially with a specific feed rate of 0.06 h 1. Correct feed
addition was obtained programming the fb-pro software (DasGip)
and controlled using the DasGip control system. An organic layer
of dodecane (Sigma-Aldrich) was added aseptically to a final
volume of 10% (v/v) immediately before starting the feed.
2.10. Analytical methods
The cell dry weight was measured by filtering known volumes
of the cultures through pre-dried and pre-weighed 0.45-mmpores size nitrocellulose filters (Supor-450 membrane filters;
PALL Life Sciences Ann Abor, MI). The filters with the biomass
were washed with water, dried for 15 min in a microwave oven at
150 W, and weighed again. The optical density at 600 nm was
determined using a Hitachi U-1100 spectrophotometer.
Concentrations of glucose, glycerol, ethanol, acetate, succinate
and pyruvate were analyzed by an isocratic high-performance
liquid chromatography (UltiMates 3000 Nano, Dionex) with an
Aminex HXP-87H ion-exchange column (Bio-Rad, Hercules, CA) at
65 1C using 5 mM H2SO4 as mobile phase at a flow rate of
0.6 ml min 1. Glucose, glycerol and ethanol were measured with
a refraction index detector (RI-101 Refractive Index Detector,
Shodexs), and acetate, succinate and pyruvate were measured
with a UV–visible light absorbance detector (UltiMate 3000
Variable Wavelength Detector, Dionex).
2.13. Analysis of total sterol fraction
For the extraction of sterols, a previously described method
(Asadollahi et al., 2010) was used with minor modifications.
Known volumes of fermentation broth were harvested by centrifuging at 5000 rpm for 10 min. The cell pellet was washed
twice with distilled water and the cell suspension was centrifuged
for another 10 min at 5000 rpm. The cell pellet was re-suspended
in 4 ml of 0.2 N HCl and heated in a water bath at 85 1C for 1 h and
then allowed to cool to room temperature. After centrifugation for
10 min at 5000 rpm and removal of the supernatant, the cell
pellet was resuspended in 2 ml methanol containing 0.2% (w/v)
pyrogallol and 1 ml 4 N KOH and transferred to a 14 ml glass vial
sealed with a PTFE lined screw cap, heated again for 2 h in a water
bath at 85 1C for saponification and then cooled to room temperature. Sterols were extracted by addition of 5 ml heptane
followed by vigorous mixing for 2 min. After 2 h, the n-heptane
layer was transferred to a new glass vial for HPLC analyses.
Quantitative determination of total ergosterol was carried out
by a isocratic high-performance liquid cromatograph (UltiMates
3000 Nano, Dionex) with a reverse phase Develosil column (C30UG-5; Nomura Chemicals, Aichi, Japan) at 40 1C using 70% MeOH
as the mobile phase at a flow rate of 1 ml min 1. The ergosterol
concentration was measured with a UV–visible light absorbance
detector set at 280 nm (Photodiode Array Detector, Dionex). The
amount of ergosterol was determined with Dionex Chromeleons
software using absolute calibration curves.
2.14. Calculation of specific rates and yield coefficients
The maximum specific growth rates, yield coefficients, specific
product formation rates and specific substrate consumption rates
are expressed as mg or g product substrate per g dry cell weight
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
6
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
per hour and were calculated as previously described (Nielsen
et al., 2003) All calculations are limited to the exponential
glucose-limited growth phase.
3. Results
3.1. a-Santalene production in S. cerevisiae
The a-santalene production in S. cerevisiae was initially evaluated introducing the expression plasmid pICK01 containing a
copy of tHMG1 and santalene synthase (SanSyn) under control of
the PGK1 and TEF1 promoter, respectively, into a lpp1D strain to
reduce farnesol formation (Faulkner et al., 1999) resulting in strain
SCICK13. Production capacity was tested cultivating the strain in a
batch in situ product removal (ISPR) reactor mode, consisting of an
aqueous two-phase partitioning system. This strategy had previously been used successfully to increase product recovery in
different microbial production processes (Daugulis, 1991; Stark
and von Stockar, 2003). Dodecane was selected as organic phase
due to its hydrophobicity (Log Pdodecane: 6.6; Log Psantalene: 6.2), low
volatility and biocompatibility with S. cerevisiae growth (Newman
et al., 2006; Asadollahi et al., 2008). Product accumulation in the
dodecane layer was monitored by gas chromatography-mass
spectrometry (GC/MS). The transformed strain was able to synthesize a low amount of a-santalene. Analysis of the organic layer
revealed a major chromatographic peak corresponding in retention
time and mass spectrum to the plant-extracted a-santalene and a
second minor peak also present in the plant extract, subsequently
identified as trans-a-bergamotene (Fig. 5). The amount of the
sesquiterpenes a-santalene and trans-a-bergamotene produced
were, respectively, 1.4570.02 and 0.1770.01 mg/l.
3.2. MET3 promoter activity during shake flask cultivation
As a next step, we wanted to increase a-santalene production by
reducing ergosterol formation. A promoter, which is widely used for
lowering the expression of ERG9 encoding squalene synthase and
hereby increasing the FPP pool, is the MET3 promoter. As mentioned
above, this might not be an optimal solution, since the repressing
agent, methionine, is metabolized by the cells thus releasing
repression. This was demonstrated by fusing PMET3 to the lacZ gene
followed by measuring b-galactosidase activity of the resulting
strain (SCICK10) cultivated in shake flasks without L-methionine
or supplied with 1 mM and 2 mM L-methionine, respectively. The
concentrations of the inhibitor were chosen based on the amounts
previously used for ERG9 repression (Asadollahi et al., 2008). As
shown in Fig. 1, the b-galactosidase activity was constant in the
cultures not containing L-methionine. In the cultures that contained
L-methionine, LacZ activity was initially very low. However, at about
mid-exponential phase, it started to increase and rapidly reached
the levels measured in the non-repressed culture.
These results thus demonstrate the difficulties controlling
promoter activity when the repressing agent is metabolized by
the cells. We therefore tested, if D-methionine or 2-hydroxy-4(methylthio)butyric acid could serve as L-methionine analogs to
repress the PMET3 promoter, because they may not be metabolized
by yeast or metabolized to a lesser extent. At concentrations of up
to 4 mM in the medium neither of the two compounds had the
capability to reduce PMET3 activity (data not shown).
3.3. Evaluation of alternative promoters for controlling squalene
synthase activity
Because of the disadvantages of the MET3 promoter, alternative systems were evaluated to down-regulate ERG9 expression.
Fig. 1. LacZ activity in strain SCICK10 (PMET3-lacZ) in response to different
methionine concentrations, 0 mM (diamonds), 1 mM (triangles) and 2 mM
(circles). Strains were cultivated in duplicates, glucose exponential growth phase
was between 2 and 16 h of cultivation.
The chosen regulatory systems were (i) the low-level constitutive
TEF1 promoter mutant PTEF1M2 selected after a directed evolution
approach based on error-prone PCR (Alper et al., 2005; Nevoigt
et al., 2006), (ii) the glucose concentration controlled promoter of
the hexose transporter gene HXT1 (Ozcan and Johnston, 1995;
Lewis and Bisson, 1991) and (iii) the HXT2 promoter potentially
useful for a gene silencing approach expressing ERG9 antisense
mRNA (Ozcan and Johnston, 1995). The approach of using
promoters, which are regulated by glucose concentration was
chosen as a means to achieve moderate expression levels during
exponential growth in batch cultivation, i.e. at high glucose
concentration and maximal repression during low glucose concentration, e.g. during the feed phase of a glucose-limited fedbatch process. To test whether the chosen promoters show
suitable activity levels compared to the native ERG9 promoter,
fusion constructs of PTEF1M2, PHXT1, PHXT2 and PERG9 with the lacZ
reporter gene were integrated into the yeast genome. Strains
SCICK06 (PHXT1), SCICK08 (PTEF1M2), SCICK09 (PHXT2) and SCICK11
(PERG9) were cultivated in shake flasks and LacZ activity was
monitored. Strain SCICK11 showed a steady LacZ activity level
throughout the cultivation (Fig. 2). b-galactosidase activity in
SCICK06 decreased with decreasing glucose concentration reaching the same level as in SCICK11 in late stationary phase, whereas
LacZ activity in SCICK09 increased. During exponential growth,
SCICK08 exhibited a very low activity, which increased slightly
during stationary phase. The different strains displayed similar
growth profiles and no differences were observed in biomass
formation.
Developing an efficient cultivation method is a key step in
designing a cost effective bioprocess. Fed-batch cultivation mode
is widely applied during industrial productions and is often a first
choice to achieve high productivity (Nielsen et al., 2003). As
previously shown, when controlled by its native promoter, ERG9
transcript levels display context dependency, i.e. the gene shows
different expression levels depending on the growth conditions
and the carbon source utilized (Kennedy et al., 1999; Kennedy,
2001). To investigate whether the selected promoters could also
be employed during a fed-batch based fermentation process, the
same strains were cultivated in the presence of glucose feed
beads thus simulating the feed phase in a glucose-limited fedbatch cultivation. Under these conditions, glucose is released
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
Fig. 2. LacZ activity in strains SCICK06 (PHXT1, filled diamonds), SCICK08 (PTEF1M2,
filled circles), SCICK09 (PHXT2, empty squares) and SCICK11 (PERG9, filled squares)
during shake flask cultivation. Strains were cultivated in duplicates, glucose
exponential growth phase was between 2 and 16 h of cultivation.
7
Fig. 4. Characterization of promoter strength during shake flask cultivation in fedbatch mode. b-galactosidase activity in strains SCICK06 (PHXT1, filled diamonds),
SCICK08 (PTEF1M2, filled circles), SCICK09 (PHXT2, empty squares) and SCICK11
(PERG9, filled squares). b-galactosidase activity is the average of values obtained
from at least three independent cultivations assayed in duplicates.
consumption (data not shown). The expression of b-galactosidase
controlled from the different promoters was assayed at regular
intervals and monitored during a cultivation period of 80 h. As in
the previous experiment, LacZ activity was steady in SCICK11
(PERG9) (Fig. 4). As expected, the highest activity was measured for
SCICK09 (PHXT2), the lowest for SCICK06 (PHXT1). Surprisingly, bgalactosidase activity in SCICK08 (PTEF1M2) drastically increased
during the cultivation finally reaching the same level as for
SCICK09 (PHXT2), which indicates that (random) mutagenesis
may turn a constitutive promoter – PTEF1 had previously shown
a constant level of expression throughout different cultivations
(Partow et al., 2010) – into a conditional promoter.
According to these results, PHXT1 appeared to be a suitable
promoter to down-regulate ERG9 expression under glucose limiting conditions, whereas PHXT2 was chosen to regulate the expression of an ERG9 antisense construct. Due to its high activity levels
at low glucose concentrations, the idea of employing PTEF1M2 for
ERG9 regulation was discarded.
Fig. 3. Shake flask cultivation in fed-batch mode showing absolute glucose release
kinetics without cells, residual glucose and ethanol concentration as well as
biomass formation. The data represented were obtained for strain SCICK03; all the
strains tested exhibited similar profiles.
from silicone elastomers with a controlled kinetics profile (Fig. 3),
which allows mimicking a fed-batch fermentation mode comparable to a regular fed-batch. The glucose release (in absence of
cells) as well as glucose concentration, cellular growth and byproduct formation during the entire cultivation were monitored.
This glucose restrained process allows a precise comparison of the
different constructs and to fully explore the potential of the
glucose sensing promoters. After an initial period of about 5 h of
adaptation, where glucose release was larger than the cellular
consumption, glucose became the limiting factor for growth and
all the glucose released from the elastomers was rapidly converted into biomass with no ethanol overflow metabolism or
formation of other by-products (Fig. 3). All strains exhibited a
similar growth profile and no significant differences in glucose
3.4. Evaluation of santalene production in fed-batch fermentation
mode
Based on the results of the promoter characterization study,
four strains were constructed. All strains carried a deletion in the
phosphatase encoding LPP1 gene to reduce the loss of FPP to
farnesol (Faulkner et al., 1999). tHMG1 and SanSyn were
expressed from a high copy number plasmid under control of
the PGK1 and TEF1 promoter, respectively. Strain SCICK13 containing the native ERG9 promoter served as a reference strain. In
strain SCICK12, the ERG9 promoter had been replaced by the HXT1
promoter. Strain SCICK14 carried an antisense DNA fragment
comprising the 50 region of ERG9 and part of its 50 UTR (Bonoli
et al., 2006; Olsson et al., 1997), whose expression was controlled
by the HXT2 promoter and which was integrated into chromosome XIII at a site providing high expression levels (Flagfeldt
et al., 2009). For comparison with previous approaches, strain
SCICK15, which carried PMET3 instead of PERG9, was used. In order
to maintain ERG9 repressed in the SCICK15 culture during the
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
8
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
Fig. 5. (A) Total ion chromatograms from GC–MS analysis of authentic standard of farnesol, a-santalene, and an extract of engineered S. cerevisiae showing peaks of
a-santalene (S), trans-a-bergamotene (B) and farnesol (F). The representative ion chromatogram referred to as yeast products was obtained during ISPR fed-batch
fermentation of strain SCICK12. (B) Mass spectra and retention times of a-santalene produced from yeast and extracted from plant (left panel) and E,E-farnesol produced
from yeast and chemical standard (right panel).
fermentation, L-methionine was added at regular intervals every
6 h to a final concentration of 2 mM.
Physiological characterization of the strains was completed in
aerobic glucose-limited fed-batch cultures (Table 3). A fed-batch
in situ product removal (ISPR) reactor mode was chosen to
evaluate the a-santalene production capacity of these strains
engineered to accumulate FPP. Cultivation was started as batch
with 30 g/l of glucose. After complete glucose consumption and
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
9
Table 3
Physiological parameters obtained during fed-batch cultivation of strains SCICK13, SCICK12, SCICK14, SCICK15 and SCICK17.
Strain
SCICK13
SCICK12
SCICK14
SCICK15
SCICK17
(PERG9)
(PHXT1)
(PHXT2)
(PMET3)
(PHXT1)
m (h 1)
YSX (g g 1)
rs (mmol (g
biomass) 1 h 1)
r CO2 (mmol
(g biomass) 1 h 1)
r O2 (mmol (g
biomass) 1 h 1)
RQ
TotSant (mg l 1)
0.0617 0.006
0.0567 0.005
0.0647 0.006
0.0577 0.006
0.0577 0.005
0.507 0.02
0.507 0.01
0.487 0.03
0.497 0.03
0.497 0.01
0.58 7 0.03
0.58 7 0.02
0.64 7 0.28
0.607 0.01
0.607 0.01
1.22 7 0.06
1.32 7 0.01
1.57 7 0.01
1.49 7 0.05
1.46 7 0.08
1.29 70.03
1.43 70.03
1.62 70.04
1.53 70.07
1.48 70.09
0.957 0.01
0.927 0.01
0.977 0.01
0.967 0.01
0.987 0.01
49.86 70.23
75.73 70.34
49.00 70.34
46.56 70.87
91.96 70.71
Specific growth rate m (h 1). Biomass yield YSX (g biomass (g substrate) 1). Specific carbon dioxide production rate (rCO2 ) and substrate (rs) and oxygen (r O2 ) consumption
rates (mmol (g biomass) 1 h 1). Respiratory quotient RQ measured as r CO2 /r O2 . Referred to the initial 30 h of the glucose-limited feed phase. TotSant, a-santalene titer
(mg l 1) measured at the end of the feed process. Values represent the mean 7 S.D. of two independent cultivations.
Fig. 6. Time course of an aerobic fed-batch culture with exponential sugar feed of S. cerevisiae strains SCICK12 (C) SCKCK14 (D) SCICK13 (E) SCICK15 (F). The feed of glucose
(ml h 1) is shown on the upper graph (A) and (B). Typical profile observed for formation of biomass (g l 1, filled diamonds); a-santalene (mg l 1, filled squares);
E,E-farnesol (mg l 1, filled cycles) carbon dioxide production CTR (mmol h 1, lines) are represented. Data represent the average of two independent cultures.
after residual ethanol produced during the glucose consumption
phase was completely depleted, the organic layer was added to
the fermentor and the production phase was started by initiating
a feed of fresh concentrated substrate with exponential kinetics
for a total feed period of 36 h (Fig. 6). Within the first 30 h of feed
the culture metabolism was completely respiratory characterized
by complete oxidation of glucose with biomass and carbon
dioxide as the major products and complete absence of fermentation products, while the respiratory coefficients remained close to
1 for all strains (Table 3). The period of respiratory growth was
followed by a phase where yeast growth was no longer consistent
with the feeding profile resulting a shift towards fermentative
metabolism accompanied by accumulation of glucose and ethanol
(data not shown). To examine the effect of ERG9 repression on the
sterol pathway, the total cellular sterol content was measured.
Both the two PERG9 replacement mutants and the strain expressing the antisense construct showed a lowered sterol content
when compared to the strain containing the original ERG9
promoter. The decrease in ergosterol ranged from 50 to 91%,
and strain SCICK12 (PHXT1) showed the lowest sterol content
(Fig. 7).
To establish if the lower sterol content reflected an increased
availability of FPP precursor for sesquiterpene conversion, product accumulation in the organic layer was measured. Similarly
to the results in shake flasks, formation of a-santalene was
accompanied by trans-a-bergamotene production observed in
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
10
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
Fig. 8. Effect of single lpp1D and lpp1D dpp1D double deletion on a-santalene, E,Efarnesol and ergosterol production rates (mg g biomass 1 h 1) in strains SCICK12
(PHXT1 lpp1D) (empty bars) and SCICK17 (PHXT1 lpp1D dpp1D) (filled bars). The error
bars represent the standard deviation for two independent cultivations.
Fig. 7. (A) Ergosterol production rate (mg g biomass 1 h 1). (B) a-santalene and
E,E-farnesol production rate (mg g biomass 1 h 1) in strains SCICK13 (PERG9),
SCICK14 (PHXT2-as), SCICK15 (PMET3) and SCICK12 (PHXT1). Strains were grown in
a two-phase partitioned fed-batch glucose-limited cultivation mode. The error
bars represent the standard deviation for two independent cultivations.
all strains. The amount of trans-a-bergamotene detected was
proportionally consistent to the amount of a-santalene in all
cases and corresponded to about 12% of the total a-santalene
produced confirming that this compound is a secondary product
of santalene synthase. A linear correlation was observed between
the reduction in ergosterol content and the increase in sesquiterpene production. Different levels of ERG9 repression diverted
FPP towards santalene formation redirecting the flux distribution
from the sterol pathway to sesquiterpene production with a
santalene productivity ranging between 0.054 and 0.183 mg
(g biomass) 1 h 1 (Figs. 6 and 7). Strain SCICK12 (PHXT1) was
the best santalene producer with a 3.4 fold improvement in
santalene productivity compared to SCICK13 (PERG9).
As previously described, ERG9 down-regulation results in accumulation of farnesol (Takahashi et al., 2007; Asadollahi et al., 2008,
2010). Substantial accumulation of farnesol, an FPP-derived byproduct, was detected in all strains (Fig. 7). Farnesol formation was
inversely correlated with the ergosterol content and correlated to
the santalene accumulation pattern observed, suggesting that ERG9
down-regulation and HMGR over-expression combined with glucose de-repressed growth conditions resulted in an increased
availability of FPP for a-santalene synthesis.
3.5. Contribution of DPP1 deletion to santalene production
DPP1 deletion had been used previously in order to reduce
farnesol accumulation in a sesquiterpene producing strain
(Takahashi et al., 2007). Together with Lpp1, Dpp1 is responsible
for most of the cytosolic isoprenoid and lipid phosphate phosphatase activity in S. cerevisiae (Toke et al., 1998; Faulkner et al.,
1999). As all engineered strains showed an increased conversion
of FPP to FOH thus reducing substrate availability for santalene
synthase the additional effect of deletion of DPP1 on the best
producing strain SCICK12 (PHXT1) was investigated. DPP1 was
deleted in strain SCICK01 and subsequently transformed with
the expression plasmid pICK01 containing a copy of tHMG1 and
santalene synthase resulting in strain SCICK17. Deletion of DPP1
resulted in an increase of the a-santalene specific production rate
from 0.18 to 0.21 mg (g biomass) 1 h 1 together with a 24% drop
in farnesol accumulation, but showed only a minor effect on the
ergosterol content (Fig. 8).
Using the different promoter systems, it was possible to
increase the santalene productivity from 0.05 to 0.18 mg
(g biomass) 1 h 1. DPP1 deletion contributed to reduce the FOH
formation and further increase the final santalene titer by 54%
(Table 3). Combining these modifications resulted in a strain
capable of the highest a-santalene production level of any strain
tested with a final titer of 92 mg/liter.
4. Discussion
Previous studies have reported successful examples of expression of different plant derived terpene synthases in the yeast
S. cerevisiae (Yamano et al., 1994; Jackson et al., 2003; Dejong
et al., 2006; Ro et al., 2006; Asadollahi et al., 2008). In this work,
yeast was engineered for the first time to produce a-santalene
introducing santalene synthase (SanSyn) from C. lansium. As
observed during expression in Escherichia coli (Schalk, 2011), asantalene was the main product formed by this enzyme and only
a minor amount of the secondary compound trans-a-bergamotene was detected. Catalytic efficiency and specificity of the
heterologous synthase are often referred to as key factors in order
to achieve relevant titers of the desired compound (Picaud et al.,
2005).
Terpene precursors are produced in yeast through the mevalonate pathway (MVA). Due to the variety of essential compounds
derived from this pathway, the activity of many enzymes is
strictly regulated at different levels (Maury et al., 2005). Yeast
was engineered to increase a-santalene production by modulating the expression of two key metabolic steps of the mevalonate
pathway (i) down-regulating the squalene synthase gene (ERG9)
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
and (ii) over-expressing the truncated version of HMG-CoA
reductase (tHMG1) to increase the pool of the critical intermediate FPP and enabling the redirection of the carbon flux towards
a-santalene.
FPP-derived squalene is a critical precursor of ergosterol, a key
component of the yeast cytoplasmic membrane and essential for
membrane fluidity. The attempt to increase the FPP pool by ERG9
deletion resulted in a complete loss of squalene formation and has
been shown to be lethal (Jennings et al., 1991). Ergosterol
supplementation for restoring viability would be economically
unfeasible for industrial applications. Recently, many different
techniques have been applied to reduce a specific gene activity as
a suitable alternative to complete gene deletions (Mijakovic et al.,
2005; Hammer et al., 2006). In this work, several strategies for
down-regulation of ERG9 were evaluated to precisely adjust
enzyme activity throughout the entire course of fermentation
enabling sufficient squalene production to fulfill the minimum
ergosterol requirements to sustain cellular growth without extracellular sterol supplementation and improve FPP availability for
conversion into a-santalene. Characterization of the promoter
activity based on a lacZ gene reporter assay allowed the identification of promoters that could provide the optimal level of SQS
necessary to optimize a-santalene production. Based on their
activity profiles, PHXT1 and PHXT2 were chosen to promote expression of ERG9 and antisense ERG9, respectively, in a fed-batch
process with the aim to couple SQS activity to glucose concentration, i.e. to achieve maximal repression during the feed phase
when glucose is limiting. Among the different systems tested,
repression of ERG9 transcription under glucose limitation using
the PHXT1 promoter was more efficient than induction of antisense
RNA controlled by PHXT2 or employing the previously used PMET3
promoter. The minor effect observed using the antisense strategy
is probably due to the fact that the expression of asRNA resulted
in incomplete gene repression in yeast (Bonoli et al., 2006). Using
a glucose responsive promoter has the additional advantage that
no addition of an expensive repressing or inducing agent is
needed to control its activity.
Data reported in this work show that engineering the FPP
branch point increases the cellular pool of FPP reducing the sterol
content and leading to an effective enhancement of flux towards
sesquiterpenes. Applying different levels of repression of SQS
resulted in a consistent redirection of carbon from ergosterol
towards a-santalene and FOH. A linear correlation was observed
between the different levels of decrease in sterol content and
sesquiterpene (santalene þfarnesol) formation indicating that
down-regulation of ERG9 changed the availability of FPP and
resulted in diverting the flow to the sesquiterpene compounds. It
has been hypothesized that SQS has a lower affinity for FPP
compared to most of the other enzymes acting at the FPP branch
point resulting in high flux toward the non-sterol branches at low
FPP concentration (Scheffler, 2002). The catalytic performance of
santalene synthase was sufficient to compete with the non-sterol
branches and efficiently drained the FPP pool towards the
sesquiterpene compound a-santalene. However, santalene
synthase was not able to completely convert the excess of FPP
and this resulted in accumulation of small amounts of farnesol.
This hypothesis was supported by a small increase in the
ergosterol level observed when farnesol production was further
reduced by deleting DPP1 leaving the level of a-santalene mainly
unchanged. This suggests that the catalytic capacity of santalene
synthase could be saturated and therefore not sufficient to
convert the additional FPP created by down-regulation of SQS
thus resulting in FOH overflow.
Sterol alteration had no effect on the growth characteristics of
the engineered strains probably due to the reduced growth rate
(m ¼0.06 70.01) applied during the fed-batch process, which is
11
far below the maximum specific growth rate measured for theses
strains under normal batch conditions (mmax ¼0.3570.01). The
accumulation of sterol intermediates is known to result in feedback inhibition of the MVA pathway (Maury et al., 2005). Here,
the decrease in sterol content achieved by down-regulating ERG9
could contribute to relieve this regulatory mechanism and further
enrich the flow through the MVA pathway leading to high yields
of a-santalene.
According to previous reports, farnesol accumulation was
observed in strains over-expressing HMGR (Ohto et al., 2009,
2010). Conversion of FPP to farnesol is the preferred alternative
route when squalene synthase is inhibited in mouse, rat and dog
(Bansal and Vaidya, 1994). Yeast strains blocked at squalene
synthase require ergosterol for growth and produce farnesol
(Song, 2003). FOH accumulation was previously detected in yeast
strains treated with zaragozic acid, a natural inhibitor of SQS
(Kuranda et al., 2010). Moreover, the acitivity of HMGR is
increased in glucose de-repressed fermentation mode (Quain
and Haslam, 1979). Farnesol formation could be explained by
the effects of the deregulation of HMGR combined with glucose
de-repressed growth conditions that increased the intracellular
FPP concentration and shunted the FPP pool towards farnesol via
dephosphorylation. Due to the potentially toxic effect of intracellular FPP accumulation (Bansal and Vaidya., 1994), dephosphorylation could act as self-defense mechanism diverting the excess
of FPP into FOH that can then be secreted.
In contrast to other organisms where farnesol production is
attributed to specific farnesyl pyrophosphatases (Christophe and
Popja, 1961; Bansal and Vaidya, 1994), yeast enzymatic activities
involved in the FPP dephosphorylation process have not been
fully elucidated yet. Several mechanisms have been suggested,
among them (i) self-de-phosphorylation by FPP synthase (Erg20)
(Chambon et al., 1990); (ii) non-identified specific phosphatase or
pyrophospatase activities (Chambon et al., 1990) and (iii) acid
catalyzed non-enzymatic hydrolysis (Muramatsu et al., 2008).
Biochemical enzymatic characterizations demonstrated that diacylglycerol pyrophosphate phosphatase encoded by DPP1 has
broad substrate specificity and can utilize isoprenoid phosphate
compounds as substrate (Faulkner et al., 1999; Carman and Wu,
2007). Deletion of DPP1 has been previously used in an attempt to
reduce dephosphorylation of FPP to FOH during isoprenoid
production resulting in a reduction of 67% in the FOH production
from 90 mg/l to 30 mg/l (Takahashi et al., 2007). Together with
Dpp1, lipid phosphate phosphatase Lpp1 accounts for most of the
cytosolic lipid phosphate phosphatase activity in S. cerevisiae
(Toke et al., 1998). When two enzymes compete for the same
substrate the catalytic efficiency (Vmax/Km) may represent a
decisive parameter for increasing the flux through a specific
enzyme. Previous work showed that simultaneous knock-out of
LPP1 and DPP1 reduced the rate of hydrolysis of FPP into FOH
in vitro to about 10% (Faulkner et al., 1999). Introducing DPP1
deletion in an lpp1D strain further improved a-santalene productivity in the fermentation process at the expense of FOH
formation.
An efficient fermentation process strategy that couples biochemical production to biomass formation was utilized for
improving the production of a-santalene. Fed-batch fermentation
operations are commonly used during industrial production
processes to achieve a high yield and productivity of the target
product (Nielsen et al., 2003). Limited exponential feed profiles of
glucose for S. cerevisiae fed-batch cultivations were used to
maximize the carbon flux from glucose to biomass and the
desired target compound, alleviating glucose repression and
Crabtree effect (Pronk et al., 1996). Due to the low water
solubility of a-santalene the compound easily gets stripped with
the gas bubbles used for aeration, but here the product was
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
12
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
captured using an in situ product removal bioreactor set-up. This
technique has been intensively used to enhance the production of
high value products such as secondary metabolites (Daugulis,
1997). This double phase partitioning system allows an in situ
product capturing in the bioreactor minimizing loss of volatiles
and compounds with low solubility through the gas outlet and
reducing potential toxic effect due to product accumulation.
Combining a metabolic engineering strategy together with
fermentation optimization, a production process capable of reaching industrial relevant amounts of the compound a-santalene was
realized. Final titers of approximately 92 mg l 1 and 131 mg l 1
of a-santalene and total sequiterpene were reached in 36 h of
feed from a synthetic minimal medium. Furthermore, our study
shows that through the use of glucose concentration regulated
promoters it is possible to dynamically redirect carbon fluxes in
the cell during fed-batch fermentation, and this approach may
find application also in the production of a wide range of other
products by yeast.
Acknowledgments
Part of this work has been financed by the Knut and Alice
Wallenberg Foundation and the Chalmers Foundation. The
authors would like to thank Anne Pedersen for her valuable
assistance in the strain construction.
References
Ajikumar, P.K., Tyo, K., Carlsen, S., Mucha, O., Phon, T.H., Stephanopoulos, G., 2008.
Terpenoids: opportunities for biosynthesis of natural product drugs using
engineered microorganisms. Mol. Pharmacol. 5, 167–190.
Alper, H., Fischer, C., Nevoigt, E., Stephanopoulos, G., 2005. Tuning genetic control
through promoter engineering. Proc. Natl. Acad. Sci. USA 102, 12678–12683.
Asadollahi, M.A., Maury, J., Møller, K., Nielsen, K.F., Schalk, M., Clark, A., Nielsen, J.,
2008. Production of plant sesquiterpenes in Saccharomyces cerevisiae: effect of
ERG9 repression on sesquiterpene biosynthesis. Biotechnol. Bioeng. 99,
666–677.
Asadollahi, M., Maury, J., Patil, K.R., Schalk, M., Clark, A., Nielsen, J., 2009.
Enhancing sesquiterpene production in Saccharomyces cerevisiae through in
silico driven metabolic engineering. Metab. Eng. 11, 328–334.
Asadollahi, M.A., Maury, J., Schalk, M., Clark, A., Nielsen, J., 2010. Enhancement of
farnesyl diphosphate pool as direct precursor of sesquiterpenes through
metabolic engineering of the mevalonate pathway in Saccharomyces cerevisiae.
Biotechnol. Bioeng. 106, 86–96.
Baldovini, N., Delasalle, C., Joulain, D., 2010. Phytochemistry of the heartwood
from fragrant Santalum species: a review. Flavour Fragr. J. 26, 7–26.
Bansal, V.S., Vaidya, S., 1994. Characterization of two distinct allyl pyrophosphatase activities from rat liver microsomes. Arch. Biochem. Biophys. 315,
393–399.
Basson, M.E., Thorsness, M., Rine, J., 1986. Saccharomyces cerevisiae contains two
functional genes encoding 3-hydroxy-3-methylglutaryl-coenzyme A reductase. Proc. Natl. Acad. Sci. USA 83, 5563–5567.
Bonoli, M., Graziola, M., Poggi, V., Hochkoeppler, A., 2006. RNA complementary to
the 50 UTR of mRNA triggers effective silencing in Saccharomyces cerevisiae.
Biochem. Biophys. Res. Commun. 339, 1224–1231.
Brown, M.S., Goldstein, J.L., 1980. Multivalent feedback regulation of HMG CoA
reductase, a control mechanism coordinating isoprenoid synthesis and cell
growth. J. Lipid Res. 21, 505–517.
Carman, G.M., Wu, W.I., 2007. Lipid phosphate phosphatases from Saccharomyces
cerevisiae. Methods Enzymol. 434, 305–315.
Chambon, C., Ladeveze, V., Oulmouden, A., Servouse, M., Karst, F., 1990. Isolation
and properties of yeast mutants affected in farnesyl diphosphate synthetase.
Curr. Genet. 18, 41–46.
Chang, M.C., Keasling, J.D., 2006. Production of isoprenoid pharmaceuticals by
engineered microbes. Nat. Chem. Biol. 2, 674–681.
Christophe, J., Popja, G., 1961. Study on biosynthesis of cholesterol: XIV. The origin
of prenoic acids from allyl pyrophosphates in liver enzyme systems. J. Lipid
Res. 2, 244–257.
Corey, E.J., Chow, S.W., Scherrer, R.A., 1957. The synthesis of a-santalene and of
trans-D11,12-iso-a-santalene. J. Am. Chem. Soc. 79, 5773.
Daugulis, A.J., 1991. Integrated product formation and recovery. Curr. Opin.
Biotechnol. 2, 408–412.
Daugulis, A.J., 1997. Partitioning bioreactors. Curr. Opin. Biotechnol. 8, 169–174.
Daum, G., Lees, N.D., Bard, M., Dickson, R., 1998. Biochemistry, cell biology and
molecular biology of lipids of Saccharomyces cerevisiae. Yeast 14, 1471–1510.
Daviet, L., Schalk, M., 2010. Biotechnology in plant essential oil production:
progress and perspective in metabolic engineering of the terpene pathway.
Flavour Fragr. J. 25, 123–127.
Dejong, J.M., Liu, Y., Bollon, A.P., Long, R.M., Jennewein, S., Williams, D., Croteau,
R.B., 2006. Genetic engineering of taxol biosynthetic genes in Saccharomyces
cerevisiae. Biotechnol. Bioeng. 93, 212–224.
Donald, K.A., Hampton, R.Y., Fritz, I.B., 1997. Effects of overproduction of the catalytic
domain of 3-hydroxy-3-methylglutaryl coenzyme A reductase on squalene
synthesis in Saccharomyces cerevisiae. Appl. Environ. Microbiol. 63, 3341–3344.
Dwivedi, C., Guan, X., Harmsen, W.L., Voss, A.L., Goetz-Parten, D.E., Koopman, E.M.,
Johnson, K.M., Valluri, H.B., Matthees, D.P., 2003. Chemopreventive effects of
alpha-santalol on skin tumor development in CD-1 and SENCAR mice. Cancer
Epidemiol. Biomarkers Prev. 12, 151–156.
Erdeniz, N., Mortensen, U.H., Rothstein, R., 1997. Cloning-free PCRbased allele
replacement methods. Genome Res. 7, 1174–1183.
Farhi, M., Marhevka, E., Masci, T., Marcos, E., Eyal., Y., Ovadis, M., Abeliovich, H.,
Vainstein, A., 2011. Harnessing yeast subcellular compartments for the
production of plant terpenoids. Metab. Eng. 13, 474–481.
Faulkner, A., Chen, X., Rush, J., Horazdovsky, B., Waechter, C.J., Carman, G.M.,
Sternweis, P.C., 1999. The LPP1 and DPP1 gene products account for most of
the isoprenoid phosphate phosphatase activities in Saccharomyces cerevisiae. J.
Biol. Chem. 274, 14831–14837.
Flagfeldt, D.B., Siewers, V., Huang, L., Nielsen, J., 2009. Characterization of
chromosomal integration sites for heterologous gene expression in Saccharomyces cerevisiae. Yeast 26, 545–551.
Förster, J., Famili, I., Fu, P., Palsson, B.Ø., Nielsen, J., 2003. Genome-scale reconstruction of the Saccharomyces cerevisiae metabolic network. Genome Res. 13,
244–253.
Goldstein, J.L., Brown, M.S., 1990. Regulation of the mevalonate pathway. Nature
343, 425–430.
Güldener, U., Heck, S., Fiedler, T., Beinhauer, J., Hegemann, J.H., 1996. A new
efficient gene disruption cassette for repeated use in budding yeast. Nucleic
Acids Res. 24, 2519–2524.
Hammer, K., Mijakovic, I., Jensen, P.R., 2006. Synthetic promoter libraries-tuning of
gene expression. Trends Biotechnol. 24, 53–55.
Hampton, R.Y., Rine, J., 1994. Regulated degradation of HMG-CoA reductase, an
integral membrane protein of the endoplasmic reticulum, in yeast. J. Cell Biol.
125, 299–312.
Hampton, R., Dimster-Denk, D., Rine, J., 1996. The biology of HMG-CoA reductase:
the pros of contra-regulation. Trends Biochem. Sci. 21, 140–145.
Howes, M.J., Simmonds, M.S., Kite, G.C., 2004. Evaluation of the quality of
sandalwood essential oils by gas chromatography-mass spectrometry.
J. Chromatogr. A 1028, 307–312.
Jackson, B.E., Hart-Wells, E.A., Matsuda, S.P., 2003. Metabolic engineering to
produce sesquiterpenes in yeast. Org. Lett. 5, 1629–1632.
Janssens, L., De Pooter, H.L., Shamp, N.M., Vandamme, E.J., 1992. Production of
flavours by microorganisms. Process Biochem. 27, 195–215.
Jennings, S.M., Tsay, Y.H., Fisch, T.M., Robinson, G.W., 1991. Molecular cloning and
characterization of the yeast gene for squalene synthetase. Proc. Natl. Acad.
Sci. USA 88, 6038–6042.
Jeude, M., Dittrich, B., Niederschulte, H., Anderlei, T., Knocke, C., Klee, D., Büchs, J.,
2006. Fed-batch mode in shake flasks by slow-release technique. Biotechnol.
Bioeng. 95, 433–445.
Julia, M., 1976. Process for Synthesizing Cis-a-Santalene and Cis-a-Santalol. US
Patent No. 3970706.
Kennedy, M.A., Barbuch, R., Bard, M., 1999. Transcriptional regulation of the
squalene synthase gene (ERG9) in the yeast Saccharomyces cerevisiae. Biochim.
Biophys. Acta 1445, 110–122.
Kennedy, M.A., 2001. Positive and negative regulation of squalene synthase
(ERG9), an ergosterol biosynthetic gene, in Saccharomyces cerevisiae. Biochem.
Biophys. Acta 1517, 177–189.
Kirby, J., Romanini, D.W., Paradise, E.M., Keasling, J.D., 2008. Engineering triterpene production in Saccharomyces cerevisiae—beta-amyrin synthase from
Artemisia annua. FEBS J. 275, 1852–1859.
Kirby, J., Keasling, J.D., 2009. Biosynthesis of plant isoprenoids: perspectives for
microbial engineering. Annu. Rev. Plant Biol. 60, 335–355.
Kuranda, K., Franc- ois, J., Palamarczyk, G., 2010. The isoprenoid pathway and
transcriptional response to its inhibitors in the yeast Saccharomyces cerevisiae.
FEMS Yeast Res. 10, 14–27.
Lewis, D.A., Bisson, L.F., 1991. The HXT1 gene product of Saccharomyces cerevisiae is
a new member of the family of hexose transporters. Mol. Cell. Biol. 11,
3804–3813.
Ma, S.M., Garcia, D.E., Redding-Johanson, A.M., Friedland, G.D., Chan, R., Batth, T.S.,
Haliburton, J.R., Chivian, D., Keasling, J.D., Petzold, C.J., Lee, T.S., Chhabra, S.R.,
2011. Optimization of heterologous mevalonate pathway through the use of
variant HMG-CoA reductases. Metab. Eng. 13, 588–597.
Maury, J., Asadollahi, M.A., Møller, K., Clark, A., Nielsen, J., 2005. Microbial
isoprenoid production: an example of green chemistry through metabolic
engineering. Adv. Biochem. Eng. Biotechnol. 100, 19–51.
Mijakovic, I., Petranovic, D., Jensen, P.R., 2005. Tunable promoters in systems
biology. Curr. Opin. Biotechnol. 16, 329–335.
Muramatsu, M., Ohto, C., Obata, S., Sakuradani, E., Shimizu, S., 2008. Various oils
and detergents enhance the microbial production of farnesol and related
prenyl alcohols. J. Biosci. Bioeng. 106, 263–267.
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
G. Scalcinati et al. / Metabolic Engineering ] (]]]]) ]]]–]]]
Nevoigt, E., Kohnke, J., Fischer, C.R., Alper, H., Stahl, U., Stephanopoulos, G., 2006.
Engineering of promoter replacement cassettes for fine-tuning of gene expression in Saccharomyces cerevisiae. Appl. Environ. Microbiol. 72, 5266–5273.
Newman, J.D., Marshall, J., Chang, M., Nowroozi, F., Paradise, E., Pitera, D., Newman, K.L., Keasling, J.D., 2006. High-level production of amorpha-4,11-diene in
a two-phase partitioning bioreactor of metabolically engineered Escherichia
coli. Biotechnol. Bioeng. 95, 684–691.
Nielsen, J., Villadsen, J., Liden, G., 2003. Bioreaction Engineering Principles, second
ed. Kluywer Plenum, New York.
Nielsen, J., 2009. Systems biology of lipid metabolism: from yeast to human. FEBS
Lett. 583, 3905–3913.
Ohto, C., Muramatsu, M., Obata, S., Sakuradani, E., Shimizu, S., 2009. Overexpression of the gene encoding HMG-CoA reductase in Saccharomyces cerevisiae for
production of prenyl alcohols. Appl. Microbiol. Biotechnol. 82, 837–845.
Ohto, C., Muramatsu, M., Obata, S., Sakuradani, E., Shimizu, S., 2010. Production of
geranylgeraniol on overexpression of a prenyl diphosphate synthase fusion
gene in Saccharomyces cerevisiae. Appl. Microbiol. Biotechnol. 87, 1327–1334.
Olsson, L., Larsen, M.E., Rønnow, B., Mikkelsen, J.D., Nielsen, J., 1997. Silencing
MIG1 in Saccharomyces cerevisiae: effects of antisense MIG1 expression and
MIG1 gene disruption. Appl. Environ. Microbiol. 63, 2366–2371.
Ozcan, S., Johnston, M., 1995. Three different regulatory mechanisms enable yeast
hexose transporter (HXT) genes to be induced by different levels of glucose.
Mol. Cell. Biol. 15, 1564–1572.
Paradise, E.M., Kirby, J., Chan, R., Keasling, J.D., 2008. Redirection of flux through
the FPP branch-point in Saccharomyces cerevisiae by down-regulating squalene
synthase. Biotechnol. Bioeng. 100, 371–378.
Partow, S., Siewers, V., Bjørn, S., Nielsen, J., Maury, J., 2010. Characterization of
different promoters for designing a new expression vector in Saccharomyces
cerevisiae. Yeast 27, 955–964.
Picaud, S., Olofsson, L., Brodelius, M., Brodelius, P.E., 2005. Expression, purification,
and characterization of recombinant amorpha-4,11-diene synthase from
Artemisia annua L. Arch. Biochem. Biophys. 436, 215–226.
Polakowski, T., Stahl, U., Lang, C., 1998. Overexpression of a cytosolic hydroxymethylglutaryl-CoA reductase leads to squalene accumulation in yeast. Appl.
Microbiol. Biotechnol. 49, 66–71.
Pronk, J.T., Steensma, H., van Dijken, J.P., 1996. Pyruvate metabolism in Saccharomyces cerevisiae. Yeast 12, 1607–1633.
Quain, D.E., Haslam, J.M., 1979. The effects of catabolite derepression on the
accumulation of steryl esters and the activity of p-hydroxymethylglutaryl-CoA
reductase in Saccharomyces cerevisiae. J. Gen. Microbiol. 111, 343–351.
13
Ro, D.K., Paradise, E.M., Ouellet, M., Fisher, K.J., Newman, K.L., Ndungu, J.M., Ho,
K.A., Eachus, R.A., Ham, T.S., Kirby, J., Chang, M.C.Y., Withers, S.T., Shiba, Y.,
Sarpong, R., Keasling, K.D., 2006. Production of the anitmalarial drug precursor
artemisinic acid in engineered yeast. Nature 440, 940–943.
Rohlin, L., Oh, M.K., Liao, J.C., 2001. Microbial pathway engineering for industrial
processes: evolution, combinatorial biosynthesis and rational design. Curr.
Opin. Microbiol. 4, 330–335.
Scallen, T.J., Sanghvi, A., 1983. Regulation of three key enzymes in cholesterol
metabolism by phosphorylation/dephosphorylation. Proc. Natl. Acad. Sci. USA
80, 2477–2480.
Schalk, M., 2011. Method for Producing Alpha-Santalene. US Pat 2011/008836 A1.
Scheffler, I.E., 2002. Metabolic Pathways Inside Mitochondria, in Mitochondria.
John Wiley & Sons, Inc., New York (pp. 298–344).
Shiba, Y., Paradise, M.E., Kirby, J., Ro, K.D., Keasling, D.J., 2007. Engineering of the
pyruvate dehydrogenase bypass in Saccharomyces cerevisiae for high-level
production of isoprenoids. Metab. Eng. 9, 160–168.
Song, L., 2003. Detection of farnesyl diphosphate accumulation in yeast ERG9
mutants. Anal. Biochem. 317, 180–185.
Stark, D., von Stockar, U., 2003. In situ product removal (ISPR) in whole cell
biotechnology during the last twenty years. Adv. Biochem. Eng./Biotechnol. 80,
150–175.
Takahashi, S., Yeo, Y., Greenhagen, B.T., McMullin, T., Song, L., Maurina-Brunker, J.,
Rosson, R., Noel, J.P., Chappell, J., 2007. Metabolic engineering of sesquiterpene
metabolism in yeast. Biotechnol. Bioeng. 97, 170–181.
Toke, D.A., Bennett, W.L., Dillon, D.A., Wu, W.I., Chen, X., Ostrander, D.B., Oshiro, J.,
Cremesti, A., Voelker, D.R., Fischl, A.S., Carman, G.M., 1998. Isolation and
characterization of the Saccharomyces cerevisiae DPP1 gene encoding diacylglycerol pyrophosphate phosphatase. J. Biol. Chem. 273, 3278–3284.
Verduyn, V., Postma, E., Scheffers, W.A., Van Dijken, J.P., 1992. Effect of benzoic
acid on metabolic fluxes in yeasts: a continuous-culture study on the
regulation of respiration and alcoholic fermentation. Yeast 8, 501–517.
Wang, C., Yoon, H.S., Jang, J.H., Chung, R.Y., Kim, Y.J., Choi, S.E., Kim, W.S., 2011.
Metabolic engineering of Escherichia coli for a-farnesene production. Metab.
Eng. doi:10.1016/j.ymben.2011.08001.
Withers, S.T., Keasling, J.D., 2007. Biosynthesis and engineering of isoprenoid small
molecules. Appl. Microbiol. Biotechnol. 73, 980–990.
Yamano, S., Ishii, T., Nakagawa, M., Ikenaga, H., Misawa, N., 1994. Metabolic
engineering for production of beta-carotene and lycopene in Saccharomyces
cerevisiae. Biosci. Biotechnol. Biochem. 58, 1112–1114.
Please cite this article as: Scalcinati, G., et al., Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the
production of plant sesquitepene a-santalene in a fed-batch mode. Metab. Eng. (2012), doi:10.1016/j.ymben.2012.01.007
PAPER II
Combined metabolic engineering of precursors and co-factor
supply to increase α-santalene production by Saccharomyces
cerevisiae.
Scalcinati G, Partow S, Siewers V, Schalk M, Daviet L, Nielsen J.
Submitted
63
1
2
3
4
Combined metabolic engineering of precursor and cofactor supply to increase α-santalene production by
Saccharomyces cerevisiae
5
Gionata Scalcinati1, Siavash Partow1, Verena Siewers1, Michel Schalk2, Laurent
6
Daviet2, Jens Nielsen1§
7
8
1
9
Technology, SE-412 96 Göteborg, Sweden.
Department of Chemical and Biological Engineering, Chalmers University of
2
Firmenich SA, Corporate R&D Division, CH-1211 Geneva 8, Switzerland
12
§
Corresponding author
13
Professor Jens Nielsen
14
Email addresses:
10
11
15
GS: [email protected]
16
SP: [email protected]
17
VS: [email protected]
18
MS: [email protected]
19
LD: [email protected]
20
JN: [email protected]
21
-1-
1
Abstract
2
3
Background
4
industrial proprieties. Due to economic difficulties of sesquiterpene production via
5
extraction process or chemical synthesis there is interest in developing alternative and
6
cost efficient bioprocesses. The hydrocarbon α-santalene is a precursor of
7
sesquiterpenes with relevant commercial applications. Here, we construct an efficient
8
Saccharomyces cerevisiae cell factory for α-santalene production.
Sesquiterpenes are a class of natural products with a diverse range of attractive
9
10
Results
11
supply was employed to manipulate the yeast metabolic network in order to redirect
12
carbon toward the desired product. To do so, genetic modifications were introduced
13
acting to optimize the farnesyl diphosphate branch point, modulate the mevalonate
14
pathway, modify the ammonium assimilation pathway and enhance the activity of a
15
transcriptional activator. The approach employed resulted in an overall α-santalene
16
yield of a 0.0052 Cmmol (Cmmol glucose)-1 corresponding to a 4-fold improvement
17
over the reference strain. This strategy, combined with a specifically developed
18
continuous fermentation process, led to a final α-santalene productivity of 0.036
19
Cmmol (g biomass)-1 h-1.
20
21
Conclusions
22
engineering strategy with fermentation technology optimization can be used to obtain
23
significant amounts of the high-value sesquiterpene α-santalene. This represents a
24
starting point toward the construction of a yeast “sesquiterpene factory” and for the
A multistep metabolic engineering strategy targeted to increase precursor and cofactor
The results reported in this work illustrate how the combination of a metabolic
-2-
1
development of an economically viable bio-based process that has the potential to
2
replace the current production methods.
3
4
5
Keywords
6
Saccharomyces cerevisiae.
7
8
Background
9
applications (e.g. flavoring agents, fragrances, food colorants, pharmaceutical agents
10
and biofuel precursors), and there has recently been much interest in biotechnological
11
production of these compounds [1-4]. Limitations in raw material accessibility, low
12
yields and high costs of the current isoprenoid production through plant extraction or
13
difficulties with chemical synthesis have caused interest in engineering cell factories
14
that can be used to produce isoprenoids in cost competitive bioprocesses [5-7].
15
Isoprenoids are natively produced in yeast though the mevalonate (MVA) pathway in
16
which the universal isoprene functional unit isopentenyl diphosphate (IPP) is
17
produced from acetyl-CoA [8]. The terminal product IPP and its isomer dimethylallyl
18
pyrophosphate (DMAPP) are subsequently condensed in the prenyl diphosphate
19
pathway generating isoprene derivatives of different chain length (C5-C20) [9]. The
20
sesquiterpene hydrocarbon α-santalene is a precursor of commercially relevant
21
sesquiterpenes (C15) and it is generated in a one-step conversion from the intermediate
22
building block farnesyl diphosphate (FPP) [10]. Stoichiometry of α-santalene (C15H24)
23
production in S. cerevisiae via the MVA pathway in purely oxidative growth
24
conditions can be summarized as:
Metabolic engineering, isoprenoids, sesquiterpenes, continuous culture,
Isoprenoids are a class of natural compounds with many potential commercial
25
-3-
1
- 4.5 C6H12O6 - 9 ATP – 6 NADPH + C15H24 + 18 NADH + 12 CO2 = 0
2
which demonstrates that α-santalene production involves a net consumption of ATP
3
and NADPH, whereas there is a net production of NADH.
4
Considerable efforts have been made to engineer yeast for isoprenoid production [8,
5
11]. Recently, progress has been reported in developing a S. cerevisiae strain capable
6
to produce commercially relevant amounts of α-santalene [12], and the aim of the
7
present work was to develop a S. cerevisiae production platform for sesquiterpene
8
compounds that could serve as an inexpensive, environmentally compatible
9
alternative to current production methods. We undertook a multistep metabolic
10
engineering strategy combining four different approaches to increase α-santalene
11
production. These included: (i) Modulation and optimization of the FPP branch point
12
(ii) De-regulation of the MVA pathway to increase the precursor pool for isoprenoid
13
synthesis (iii) Increasing the availability of the reductive cofactor NADPH by
14
modifying the ammonium assimilation pathway and (iv) Enhancing the activity of a
15
transcriptional activator of sterol biosynthesis.
16
(i) In order to minimize the overflow to the biosynthetically related sterols that have
17
the same precursor as α-santalene, FPP, the native promoter (PERG9) of squalene
18
synthase (SQS) was replaced with a glucose sensing PHXT1 promoter [12]. Previous
19
attempts to increase cytosolic FFP availability by down-regulating the ERG9 gene
20
resulted in a rapid dephosphorylation of FPP to farnesol (FOH) [13-15]. To minimize
21
the flux towards farnesol two genes, LPP1 and DPP1, encoding enzymes with FPP
22
dephosphorylation activity have been deleted [12, 16], and we also adapted this
23
approach here.
24
(ii) As a second part of the strategy we amplified the flux through the MVA pathway
25
by engineering two key enzymatic steps. The mevalonate producing 3-hydroxy-3-
-4-
1
methyl-glutaryl-CoA reductase (HMGR) enzyme is a highly regulated enzyme and is
2
generally believed to exert a high degree of flux control in the MVA pathway. Part of
3
its regulation is via the N-terminal domain of Hmg1 that spans the membrane of the
4
endoplasmic reticulum (ER) and hereby interacts with sterol sensing components of
5
the ER membrane. This feed-back regulation by sterols can be eliminated by
6
expressing a modified form of HMGR lacking the trans-membrane region [17]. Here
7
we used a genetic modification widely used in the past in order to circumvent post-
8
transcriptional regulation of HMGR [18]. The HMG1 gene region coding for the
9
catalytic domain was over-expressed resulting in a constitutively active, cytosolic
10
variant of Hmg1. This strategy has been successfully used before for over-producing
11
several different isoprenoids in S. cerevisiae [12, 14, 19-21]. The other enzymatic step
12
engineered in the MVA pathway was the one mediated by farnesyl diphosphate
13
synthase (FPPS) (encoded by the essential gene ERG20), which catalyses the
14
condensation of IPP units into geranyl diphosphate (GPP) and FPP [22]. IPP
15
condensing enzymes are interspecies conserved and the yeast ERG20 gene product
16
evolved towards specific production of FPP rather than GPP [23]. Due to the pivotal
17
nature of the FPP molecule as precursor of many essential compounds such as
18
dolichol, ubiquinone, isoprenylated proteins and ergosterol [24] its synthesis by FPPS
19
is tightly regulated and has been identified as a flux controlling step of the MVA
20
pathway, in particular controlling the intracellular FPP availability and its distribution
21
into derived products [25,26]. The efficiency of ERG20 overexpression to increase the
22
level of IPP conversion to FPP and its derivatives depends, however, on the growth
23
conditions employed and the yeast background strain utilized [20, 27]. In this study,
24
the effect of overexpressing ERG20 on α-santalene production has been investigated.
-5-
1
(iii) The manipulation of the NADH and NADPH cofactor balance in order to
2
overcome limits imposed from the cellular redox constraints is a well-established
3
metabolic engineering strategy [28]. The reaction leading to α-santalene formation
4
results in net production of NADH and consumption of NADPH (see reaction above).
5
A change in the NADH:NADPH ratio in favor of NADPH would therefore be
6
beneficial for product formation. Increasing the availability of reduced cofactor
7
NADPH by deleting the NADPH consuming reaction of glutamate dehydrogenase
8
encoded by GDH1 has previously been applied to improve product formation [29].
9
Similarly, activation of an alternative ammonium utilization route in a gdh1 strain
10
by overexpressing the NAD-dependent glutamate dehydrogenase encoded by GDH2
11
resulted in an increase of NADH consumption during the anabolic process and in a
12
modification of the yeast cofactor balance [30]. More recently, in silico analysis
13
identified the same strategy as an approach to increase sesquiterpene production [31].
14
Here we evaluated the effect of GDH1 deletion alone as well as coupled with
15
simultaneous over-expression of GDH2 on α-santalene production.
16
(iv). The last strategy we employed involved engineering of a key transcription factor
17
with the objective to generally up-regulate expression of the MVA pathway genes.
18
Upc2 and Ecm22 have been identified as the main transcription factors responsible for
19
the activation of several MVA and ergosterol pathways genes [32]. The point
20
mutation upc2-1 discovered first for conferring the ability to assimilate extracellular
21
sterols during aerobic cultivation [33] has been demonstrated to result in a
22
constitutively active form of Upc2 [34]. Overexpression of upc2-1 has been employed
23
to transcriptionally up-regulate the MVA pathway genes during isoprenoid
24
production, but its effect on enhancing the carbon flow through the pathway was
25
modest when used alone [19, 20]. However, when combined together with ERG9
-6-
1
down-regulation, it produced a clear increase in total isoprenoid production [20, 35].
2
In the current work, contribution of the upc2-1 overexpression on the production of α-
3
santalene was tested in combination with the modifications described above.
4
All the genetic modifications described above were integrated into the yeast genome
5
to enhance the genetic stability of the production strain during long term cultivation.
6
However, in order to ensure flexibility and to allow the platform strain to be used for
7
production of a range of different isoprenoids, we expressed the synthase gene
8
required for the final conversion of FPP into α-santalene together with an additional
9
copy of tHMG1 on a multicopy plasmid (Figure 1). The contribution of the different
10
metabolic engineering strategies on isoprenoid production was evaluated using an
11
integrated fermentation/downstream recovery process with a two-phase partitioning
12
continuous cultivation set-up (Figure 2). By combining the different strategies we
13
developed a yeast strain and a fermentation process that resulted in high sesquiterpene
14
titers and the results represent a first step toward the long term goal of establishing a
15
fully biotechnologically based sesquiterpene production process.
16
17
Results
18
FPP to increase the production level of the sesquiterpene α-santalene and to evaluate
19
the metabolic response of S. cerevisiae to the genetic modifications. A double phase
20
continuous cultivation method was developed as production process to investigate the
21
performances of the engineered strains at glucose-limited conditions.
22
23
24
Characterization of engineered sesquiterpene producing strains in two-phase
chemostat cultivation
25
plasmid pISP15 containing a copy of tHMG1 and codon optimized SanSyn
26
(SanSynopt) under control of the PGK1 and TEF1 promoters, respectively (strain
The primary objective of this study was to enhance the availability of intracellular
S. cerevisiae was engineered to produce α-santalene by introducing the expression
-7-
1
SCIGS28). The transformed strain was initially tested for its α-santalene producing
2
capacity in a double-phase chemostat process at a dilution rate of 0.05 h-1 resulting in
3
an α-santalene yield of 0.0013 Cmmol (Cmmol glucose)-1 and a production rate of
4
0.006 Cmmol (g biomass)-1 h-1 (corresponding to 0.086 mg (g biomass)-1 h-1). All the
5
following strain development strategies were assessed based on the yield and
6
productivity of this control strain (SCIGS28) and are reported in figure 3 and 4,
7
whereas the titers are given in figure 5.
8
Replacement of the native PERG9 promoter with PHXT1 was previously proven to
9
efficiently reduce the ergosterol production and increase the availability of FPP for
10
the conversion into sesquiterpene products [12]. Here, the same modification was
11
introduced in an lpp1Δ strain carrying the expression vector resulting in strain
12
SCIGS29. Using PHXT1 to control ERG9 expression combined with LPP1 deletion
13
resulted in an increase in α-santalene yield and productivity of 3- and 3.8-fold,
14
respectively. α-Santalene production was accompanied by the formation of the FPP-
15
derived farnesol (FOH) at a production rate of 0.006 Cmmol (g biomass)-1 h-1 (Figure
16
3A). The impact of the additional deletion of DPP1 was tested in an attempt to reduce
17
the rate of hydrolysis of FPP into the undesired by-product FOH (strain SCIGS30).
18
This resulted in an almost unchanged flux towards α-santalene formation, but in a
19
reduction of the farnesol yield and productivity by 50% and 44%, respectively.
20
In a following approach, the impact of perturbing the redox metabolism on α-
21
santalene accumulation was evaluated introducing the deletion of GDH1 encoding
22
NADP-dependent glutamate dehydrogenase (strain SCIGS31). In the strain harboring
23
the additional GDH1 deletion, no further enhancement in α-santalene productivity
24
was detected. Interestingly, no substantial FOH formation was detected in this strain
25
(Figure 3A and 4).
-8-
1
Subsequently we monitored the effect of activating an NADH consuming reaction for
2
ammonium assimilation together with the up-regulation of the second MVA pathway
3
flux controlling step FPPS, integrating and over-expressing simultaneously the genes
4
GDH2 and ERG20 (strain SCIGS24). This combination resulted in a significant
5
increase of sesquiterpene production contributing to attain the maximum α-santalene
6
yield and productivity of 0.0052 Cmmol (Cmmol glucose)-1 and 0.036 Cmmol (g
7
biomass)-1 h-1, respectively. The additional up-regulation of GDH2 and ERG20
8
combined with all previous features produced a 4- and 6-fold improvement,
9
respectively, in α-santalene yield and productivity compared to the control strain
10
(Figure 3A and 4).
11
The MVA pathway was further engineered by integrating into the yeast genome the
12
mutated transcription factor gene upc2-1 and an extra copy of tHMG1 (strain
13
SCIGS25). Previously, both strategies, using an additional genome integrated copy of
14
tHMG1 next to plasmid-based expression and the over-expression of upc2-1 have
15
displayed little or only a strain-dependent effect on final product production [14, 20].
16
Similarly, our combined approach did not contribute to increase α-santalene
17
production over the best producing strain obtained, SCIGS24. However, in contrast to
18
the insignificant change in α-santalene productivity strain SCIGS25 exhibited a 2-fold
19
increase in FOH formation yielding a final FOH yield of 0.0024 Cmmol (Cmmol
20
glucose)-1 and a productivity of 0.018 Cmmol (g biomass)-1 h-1. It is therefore worth
21
mentioning that strain SCIGS25 reached the highest total sesquiterpene yield and
22
productivity of 0.0069 Cmmol (Cmmol glucose)-1 and 0.052 Cmmol (g biomass)-1 h-1
23
(santalene + farnesol), respectively (Figure 3A and 4).
-9-
1
2
Effect of the dilution rate on sesquiterpene production
3
in the total amount of sesquiterpene produced. The sesquiterpene productivity level
4
varied almost 10-fold between the strains, from 0.006 to 0.052 Cmmol (g biomass)-1
5
h-1. Chemostat cultivation mode offers the advantage of manipulating with accuracy
6
the dilution rate that at these conditions is equal to the specific growth rate [36]. We
7
therefore decided to investigate the effect of the growth rate on sesquiterpene
8
production. All previous cultivations were performed at a dilution rate of 0.05 h-1 and
9
when the control strain was grown at D=0.1 h-1, a small decrease in the α-santalene
Under the employed conditions, the engineered strains exhibited significant changes
10
yield was observed (Figure 4) whereas its productivity remained essentially
11
unchanged (Figure 3). The increase in α-santalene production observed for strains
12
SCIGS29 and SCIGS30 at low dilution rate (D=0.05 h-1) was also seen at the higher
13
dilution rate of 0.1 h-1. α-Santalene productivities measured for these strains were,
14
respectively, 0.041 and 0.043 Cmmol (g biomass)-1 h-1 representing a 6-fold increase
15
compared to the control strain and almost double the productivity obtained when
16
growing the strains at D=0.05 h-1 (Figure 3B). In contrast, the yield was slightly
17
reduced. This showed a clear dependence of the productivity on the specific growth
18
rate applied (Figure 4). Consistently, the DPP1 deletion resulted in reduced FOH
19
accumulation in strain SCIGS30 compared to the lpp1Δ single deletion (strain
20
SCIGS29). The ratios between the α-santalene and the farnesol yield in the two strains
21
of 2.3 and 4.2, respectively, were maintained when the dilution rate was raised to 0.1
22
h-1 (Figure 4). Consistently, the same product proportion was also seen in the
23
productivities (Figure 3). Therefore, the distribution of FPP between the two products
24
appears to be independent of the specific growth rate.
25
Surprisingly, strains SCIGS31, SCIGS24 and SCIGS25 were unable to sustain growth
26
at D=0.1 h-1 and cultures were washed out (see section below).
- 10 -
1
2
Strain physiology in batch and chemostat cultivation
3
affected yeast physiology a detailed characterization of the recombinant strains was
4
carried out. Control strain SCIGS28 displayed a fully respiratory metabolism
5
(RQ=1.0) under both dilution rates. The principal physiological parameters (e.g. Ysx,
6
rs, rCO2 and rO2) were comparable with the wild type strain CEN.PK113-7D [37, 38].
7
Strains SCIGS29 and SCIGS30 exhibited a major alteration in their physiology. An
8
increase in the residual glucose concentration of 6.4 fold at D=0.05 h-1 and 2.5 fold at
9
D=0.1 h-1 was observed for both strains. As direct consequence of the increase in the
In order to evaluate if the modifications applied to increase sesquiterpene production
10
residual glucose concentration aerobic fermentation set in, resulting in ethanol
11
formation accompanied with acetate accumulation. A marked reduction in the
12
biomass yield from 0.5 to 0.29-0.28 (D=0.05 h-1) and 0.28-0.25 g biomass (g glucose)-
13
1
14
fraction corresponding to 4% (Cmmol products (Cmmol glucose)-1) of the glucose
15
consumed was fermented to ethanol and acetate. Additionally, a clear increase in the
16
glucose (rs) and oxygen consumption rate (rO2) and carbon dioxide production rate
17
(rCO2) was observed (table 3). This physiological response was observed at both
18
D=0.05 and 0.1 h-1 thus appearing to be independent of the dilution rate.
19
Despite several attempts, it was not possible to achieve a steady-state when strains
20
SCIGS31, SCIGS24 and SCIGS25 were grown at D=0.1 h-1. Instead, a progressive
21
decrease of the biomass concentration over time was observed consistent with wash-
22
out kinetics. The following characterization for these strains was therefore conducted
23
only at D=0.05 h-1.
24
When deletion of GDH1 was introduced (strain SCIGS31) a considerable fraction of
25
the glucose, 31 mmol l-1, was recovered corresponding to a consumption of only 33%
26
of the total sugar provided. However, it was still possible to reach a steady state. In
(D=0.1 h-1) was measured for the two strains (table 3). However, only a small
- 11 -
1
this strain, the rate of alcoholic fermentation increased to 0.51 mmol (g biomass)-1 h-1
2
and the metabolism shifted more predominantly to a respiro-fermentative state
3
(RQ=1.62), where 21% of the carbon source was metabolized to the fermentation
4
products ethanol and acetate. These pronounced metabolic changes were probably
5
related to the limitation in ammonium consumption as a consequence of Gdh1
6
inactivation.
7
Overexpression of GDH2 is known to partially complement the ammonium
8
assimilation defect in a gdh1Δ strain [30] and resulted in a clear reduction of the
9
ethanol and acetate production rate in strains SCIGS24 and SCIGS25 compared to
10
strain SCIGS31. It is worth noticing that in strain SCIGS24, a large decrease in the
11
biomass yield occurred and the specific glucose and O2 consumption rates and the
12
CO2 production rate increased respectively to a value of 1.16, 3.21 and 3.85 mmol (g
13
biomass)-1 h-1. The previously described overflow metabolism phenomenon towards
14
fermentation products was also observed in strains SCIGS24 and SCIGS25 and led to
15
a fraction of carbon fermented to ethanol and acetate close to 6% for both strains. All
16
engineered strains except the control strain exhibited overflow metabolism under the
17
tested conditions. The fraction of glucose converted into fermentation products ranged
18
between 0.04 and 0.21 Cmmol products (Cmmol glucose)-1. If strain SCIGS31 (that
19
exhibited a behaviour different from all other strains probably related to the major
20
role played by the ammonium limitation) is excluded from this consideration, it is
21
interesting to notice that the ratios of the different fermentation products measured
22
vary substantially between the strains. A significantly higher ethanol:acetate ratio was
23
observed for strains SCIGS24 and SCIGS25 compared to SCIGS29 and SCIGS30
24
indicating a redistribution of flux around the pyruvate dehydrogenase (PDH) bypass
25
at the acetaldehyde level. The increased ethanol:acetate ratio was reflected in an
- 12 -
1
increase in the formation of sesquiterpene products, which are derived directly from
2
the cytosolic acetyl-CoA produced through the pyruvate decarboxylase route. On the
3
other hand, the engineered strains showed a clear decrease in biomass yield compared
4
to the control strain suggesting a carbon flux redirection towards other products. The
5
fraction of carbon lost in the drop of biomass yield could not be accounted for in the
6
residual unconsumed glucose or in the fermentation products. Instead, carbon dioxide
7
was the main carbon product. Surprisingly, the increase in ethanol and acetate
8
productivity was not related to any decrease in the respiration rate. The oxygen uptake
9
rate was increased in all engineered strains compared to the control strain and reached
10
the highest value of 4.41 mmol (g biomass)-1 h-1 in strain SCIGS25 suggesting a
11
strong reprogramming of cell metabolism in these strains.
12
13
Discussion
14
aimed at increasing the production of the commercially relevant sesquiterpene
15
compound α-santalene. The strain improvement strategy was combined with
16
development of a cost effective fermentation process based on a two-phase
17
continuous cultivation mode.
18
19
Double-phase chemostat as a tool to study metabolically engineered strains
20
offer several advantages compared to batch conditions [36]. One is that they allow a
21
precise comparison of productivities of selected genetically engineered strains under
22
well-controlled constant conditions and to explore the effect of the growth rate
23
independently of the other parameters.
24
Being extensively used in bioprocesses to produce aroma compounds, in situ product
25
removal (ISPR) (for review see [39]) was applied in this study to maximize the
In this study, we provide an example of several rounds of metabolic engineering
Continuous cultivation modes have been employed in industrial bioprocesses and
- 13 -
1
product recovery. Through the combination of ISPR with chemostat cultivations we
2
obtained a production system that offers the advantage of continuous recovery of the
3
product in the fermenter effluent from the selected organic phase which can
4
subsequently be recycled, regenerated and reused in the same process for a prolonged
5
time of cultivation (for review see [40]) The developed set-up is a suitable approach
6
to develop an upscaled industrial process.
7
8
Influence of the genetic modifications on strain productivity
9
combinations on α-santalene productivity and yield. The control strain was minimally
Here we examined the impact of different metabolic engineering strategies and their
10
engineered to produce α-santalene functionally expressing a codon optimized
11
santalene syntase (SanSyn) from C. lansium and a truncated version of 3-hydroxyl-3-
12
methyl-glutaryl-coenzyme A reductase (HMGR). SanSyn belongs to the class I group
13
of sesquiterpene cylcases. These enzymes catalyze a complex intramolecular
14
cyclization of FPP with very different product specificity and the reaction mechanism
15
often involves several partial reactions [41]. Many studies have reported examples of
16
heterologous production of isoprenoids simply expressing the plant synthase in the
17
desired microbial host. However, the yields obtained are often extremely low [15, 19,
18
20, 42-45]. Similar to our previous study we decided to construct a reference α-
19
santalene producing strain (SCIGS 29) combining the synthase expression with the
20
expression of the deregulated form of Hmg1 (tHmg1) [12]. The use of tHMG1
21
represents an excellent example of bypassing one of the regulatory mechanisms
22
controlling the MVA pathway flux and has been successfully applied in a number of
23
microbial isoprenoid production processes [14, 19-21, 43]. The yield obtained in this
24
control strain was comparable with our previously reported values obtained during a
25
fed-batch process [12] and demonstrates the feasibility and robustness in applying our
- 14 -
1
novel double-phase continuous cultivation. In order to improve the production of the
2
target compound it is necessary to overcome the regulatory mechanisms that have
3
evolved to prevent flux imbalances. In this work, we modulated some of the well-
4
recognized key points that tightly regulate the carbon flux to sesquiterpenes in S.
5
cerevisiae. A slight reduction in yield and unchanged productivity observed in the
6
control strain at a higher dilution rate suggests a limitation of the plant synthase in
7
efficiently draining the FPP precursor from the MVA pathway, consistent with the
8
previous hypothesis that at low FPP concentration SantSyn competes with the other
9
cellular FPP consuming reaction [12]. A general strategy extensively applied in
10
sesquiterpene bioprocess development [14, 15, 20, 35] consists in down-regulating
11
SQS to increase the intracellular FPP pool. Replacement of the native PERG9 promoter
12
with the glucose-sensing PHXT1 promoter was recently successfully employed to divert
13
the carbon flux to sesquiterpene products instead of sterols [12]. Applying the same
14
ERG9 modification in this study together with deletion of LPP1 greatly increased the
15
sesquiterpene productivity and yield under chemostat conditions compared to the
16
control strain. The obtained productivity level appears to be linearly correlated with
17
the dilution rate employed pointing to a direct relation between the specific growth
18
rate and the overall flux through the MVA pathway and indicating that the efficiency
19
of the ERG9 modification in the enhanced FPP availability was supported at different
20
specific growth rates. A similar growth dependent relation has been reported for the
21
cellular content of ergosterol [46], which is also derived from FPP.
22
In the lpp1Δ and lpp1Δ/dpp1Δ mutants known to exhibit lower FPP phosphatase
23
activity [16, 47], the excess of FPP was redistributed between α-santalene and FOH in
24
a consistent ratio when different dilution rates were applied. These results suggest the
25
hypothesis that once a threshold level of intracellular flux toward FPP is reached the
- 15 -
1
thermodynamically favourable endogenous dephosphorylation starts and competes
2
with the catalytic capacity of santalene synthase leading to FOH accumulation. On the
3
other hand, the unchanged α-santalene yield coupled with higher productivity
4
achieved at higher dilution rates suggests that the santalene synthase was not fully
5
saturated at low dilution rates and there was excess activity to cope with high FPP
6
flux. This points out that the FOH formation is not only a direct consequence of
7
limited santalene synthase activity but that other cellular mechanisms are likely to be
8
involved. Reduction but not complete inhibition of FOH formation in the lpp1Δ
9
dpp1Δ double deletion strain compared to the single lpp1Δ deletion strain was
10
consistent with our previous report [12], and confirmed that the DPP1 encoded lipid
11
phosphate phosphatase has a role in FPP dephosphorylation and together with Lpp1 is
12
involved in the conversion of FPP into FOH. However, these are clearly not the only
13
mechanisms responsible for this conversion as we still observed some FOH
14
production in the double deletion strain.
15
Stoichiometry of the pathway reaction for α-santalene formation from glucose reveals
16
a net consumption of 0.4 mol of NADPH and net production of 1.2 mol of NADH per
17
Cmol of α-santalene formed. This fact renders the sequiterpene production pathway a
18
target for cofactor engineering to improve its productivity. Improving the NADPH
19
availability by modifying the ammonium assimilation pathway has proven to be an
20
effective strategy to increase sesquiterpene production [14]. Interestingly, when the
21
previously employed deletion of GDH1 to manipulate the cell redox metabolism was
22
introduced a reduction in α-santalene productivity without FOH accumulation was
23
obtained. This modification also strongly affected the strain physiology (see below).
24
Therefore, it is likely that the limitation in ammonium assimilation imposed by the
25
GDH1 deletion reduces the flux through the MVA pathway below the level necessary
- 16 -
1
to trigger FOH formation and conversion of FPP into α-santalene was sufficient to
2
avoid intracellular FPP accumulation.
3
Combining the simultaneous overexpression of the NAD-dependent glutamate
4
dehydrogenase and prenyl transferase encoded, respectively, by GDH2 and ERG20
5
positively affected sesquiterpene production. Overexpression of GDH2 is known to
6
restore the ammonium assimilation and consequently alter the NADH:NADPH
7
equilibrium favouring the NADPH availability at the expense of NADH produced [
8
14, 30].
9
The consensus binding motif for the sterol biosynthesis activating transcription factor
10
Upc2 has been found in most of the promoters of the ergosterol pathway genes [35].
11
Moreover, it was shown that some genes of the MVA pathway including ERG8,
12
ERG12, ERG13, ERG20 and HMG1 contain sequences similar to the consensus
13
binding sequence [32, 48]. Expression of upc2-1 together with an additional copy of
14
tHMG1 contributed to increase the carbon flux through the MVA pathway and had a
15
beneficial effect on the total sesquiterpene production. The fraction of FOH produced
16
was almost double in this strain and largely contributed to the observed increase of
17
total sesquiterpenes indicating that when the flux toward sesquiterpene is altered
18
through the introduction of genetic modifications the FPP branch point displayed an
19
unexpected flexibility in product distribution.
20
The optimal solution was obtained through combining all the modifications resulting
21
in the highest sesquiterpene yield (strains SCIGS24 and SCIGS25). Compared to our
22
previous study [12] the engineering strategy employed here allows an increase of 1.8
23
fold in α-santalene final yield (Cmmol α-santalene /Cmmol glucose). These results
24
highlight the importance of a systematic approach to achieve the ultimate goal of and
25
economically feasible sesquiterpene microbial production. It is noteworthy that
- 17 -
1
comparable sequiterpene productivity was achieved in the strains not fully engineered
2
simply by increasing the operational dilution rate (strains SCIGS29 and SCIGS30)
3
whereas the fully engineered strains were washed out when the same conditions were
4
imposed. Further studies are necessary to elucidate the factors leading to the inability
5
of these mutants to sustain growth at higher dilution rates.
6
7
Influence of genetic modifications on strain physiology
8
synthesis towards sequiterpene production by modifying the ERG9 promoter has been
9
investigated during aerobic chemostat glucose limited cultivation conditions. The
In this study, the effect of controlling the diversion of carbon flow from sterol
10
lpp1Δ and lpp1Δ/dpp1Δ mutants carrying the PHXT1-ERG9 construct clearly showed
11
an increase in the residual glucose concentration slightly above the critical
12
concentration that triggers aerobic fermentation, which was reported to lie between
13
0.5 and 0.8 mM [49, 50] and results in a typical Crabtree response. It is possible that
14
regulating the Erg9 activity using the PHXT1 glucose sensitive promoter under strictly
15
glucose limited conditions resulted in its almost complete down-regulation and in an
16
increased biosyntheic demand of the essential compound ergosterol. Ergosterol is the
17
main sterol present in the plasma membranes where it has several essential functions
18
[51]. Yeast is dependent on oxygen for sterol and fatty acid formation. Under strictly
19
anaerobic conditions this compound has to be provided in the media. Reducing its
20
provision results in a decrease of biomass formation and an increase in ethanol
21
formation [52]. Activity of PHXT1 has been shown to be induced at an extracellular
22
glucose concentration of 5.6 mM [53] suggesting that the observed increase in the
23
residual glucose concentration in the cultures was necessary to restore a minimal
24
PHXT1 activity in order to maintain the ergosterol level necessary to sustain cell
25
growth. The response to the limitation in the essential compound ergosterol could be
- 18 -
1
the reason leading to the observed decrease in biomass yield and increase of the
2
fermentative metabolism. A similar phenomenon in fact was observed in autotrophic
3
yeast strains in uracil-limited chemostat culture [54]. The observed overflow
4
metabolism toward ethanol and acetate formation increases the carbon flux through
5
the PDH bypass possibly resulting in an increase in the cytosolic acetyl-CoA
6
availability that was subsequently more effectively channelled towards the MVA
7
pathway in the engineered strains enhancing the final sesquiterpene production.
8
Strain SCIGS31 exhibited a particular physiology and needs to be discussed
9
separately. Deletion of GDH1 is known to impair the ammonium assimilation
10
resulting in a lower specific biomass formation rate on different carbon sources
11
(glucose/galactose) and under different growth conditions (batch/chemostat and
12
aerobic/anaerobic) [29, 30], which was confirmed in this study. When deletion of
13
GDH1 was introduced, ethanol formation as well as glucose accumulation occurred,
14
resulting in a situation similar to cultivation limited in essential nutrients [54]. Most
15
likely, the combination of the limitation in ammonium assimilation as result of the
16
GDH1 deletion together with the possible ergosterol limitation due to the ERG9
17
downregulation produced the observed respiro-fermentative metabolism.
18
19
Conclusions
20
pathway engineering and fermentation technologies have a significant impact in
21
accomplishing the aim to develop an economically viable biobased industrial process.
22
In this study, engineering different pathways simultaneously resulted in a robust S.
23
cerevisiae production host capable of efficiently producing α-santalene. The
24
engineered strains were evaluated in an optimized double-phase continuous
25
fermentation method leading to a high yield of α-santalene and resulting in a robust
Microbial production of sesquiterpenes is an active research area; advances in
- 19 -
1
production process that could possibly be used for commercial applications. Levels of
2
products observed open up to the possibility to explore new engineering option for
3
increasing the sesquiterpene productivity. The presented systematic metabolic
4
engineering approach represents a gateway toward the creation of yeast platform that
5
can be applied to the production of an array of sesquiterpene products.
6
Methods
7
8
Plasmid construction
9
detailed maps of the plasmids is contained in supplementary file 3. The gene coding
An overview of the plasmids constructed in this study is reported in table 1, the
10
for α-santalene synthase (SanSynopt) was codon optimized for expression in S.
11
cerevisiae and synthesized by DNA 2.0 (Menlo Park, CA, USA) (supplementary file
12
2), cut with NotI/PacI and ligated into NotI/PacI restricted vector pICK01 containing
13
tHMG1 [12] resulting in plasmid pISP15 (Figure 1).
14
To simultaneously integrate multiple genes into the yeast genome a series of plasmids
15
containing the genes, constitutive strong promoters, terminators, marker gene
16
sequences and the required region for genomic integration were constructed. All
17
endogenous S. cerevisiae genes were PCR amplified using genomic DNA of strain
18
CEN.PK113-5D as template. Primers used for amplification are provided in
19
supplementary file 1. All PCRs were performed using high fidelity PhusionTM DNA
20
polymerase (Finnzymes, Vantaa, Finland). The ERG20 gene [GenBank:
21
NM_001181600] was amplified using primers pair 1/2, subsequently digested with
22
BamHI/NheI and ligated into the vector pSP-GM2 [55] restricted with the respective
23
enzymes downstream of the TEF1 promoter resulting in plasmid pIGS01. A 711 bp
24
upstream flanking region (AD1) selected for genomic integration was amplified using
25
primers pair 3/4, cut with MreI/Kpn2I and ligated into vector pIGS01 restricted with
- 20 -
1
the respective enzymes resulting in plasmid pIGS02. Plasmid pIGS03 was obtained
2
by cloning gene GDH2 [GenBank: NM_001180275] amplified with primers pair 5/6
3
into pIGS02 downstream of the PGK1 promoter using PacI/NotI restriction sites. A
4
downstream flanking region of 653 bp (AD2) was amplified with primers 7/8,
5
digested with AscI/AvrII and ligated into pIGS03. The resulting plasmid was named
6
pIGS04. To complete the plasmid for integration the Kluyveromyces lactis (Kl) URA3
7
gene [GenBank: Y00454] was amplified with primers 9 and 10 using plasmid
8
pWJ1042 [56] as template, cut with FseI and ligated into pIGS04 after restriction with
9
the respective enzyme. The resulting plasmid was designated pIGS05, digested with
10
MreI/AscI and the resulting fragment used for integration into the yeast genome as
11
described below. The 5´ region of the Kl URA3 gene was amplified with primers 11
12
and 12, cut with AvrII/AscI and cloned into pIGS03 restricted with the respective
13
enzymes resulting in plasmid pIGS06. Amplification of the catalytic domain of the
14
HMG-CoA reductase gene (tHMG1) [GenBank: NM_001182434] was performed
15
using primers pair 13/14, the resulting fragment cleaved with NheI/BamHI and cloned
16
downstream of the TEF1 promoter into NheI/BamHI restricted pSP-GM2 resulting in
17
pIGS07. A mutant allele upc2-1 of the UPC2 gene [GenBank: NC_001180521] was
18
created by use of primer pair 15/16. To introduce the pleiotropic mutation G888D, the
19
corresponding codon GGT was mutated to GAT generating the amino acid
20
substitution. Subsequently, the PCR amplified upc2-1 was cloned downstream of the
21
PGK1 promoter into pIGS07 using NotI/PacI resulting in plasmid pIGS08. An 829 bp
22
downstream flanking region (AD3) selected for genomic integration was amplified
23
using primer pair 17/18 cut with MreI/Kpn2I and ligated into vector pIGS08 restricted
24
with the respective enzymes resulting in plasmid pIGS09. The 3´ region of Kl URA3
25
(overlapping with the 5´region described above) was amplified with primers 19 and
- 21 -
1
20, cut with AvrII/AscI and cloned into pIGS09 restricted with the respective enzymes
2
resulting in plasmid pIGS10. All plasmids were verified by sequencing (Sigma-
3
Aldrich, St. Luis, MO). Subsequently, plasmids pIGS06 and pIGS10 were restricted
4
with MreI/AscI, the cassettes isolated from the vector backbone and used for yeast
5
transformation (see below).
6
7
Yeast strain construction
8
auxotrophy for uracil [57] and are listed in table 2.
9
Strain SCIGS03 carrying a GDH1 [GenBank: NC_001183795] deletion was created
10
from strain SCICK16 using a bipartite gene-targeting technique [55]. Upstream and
11
downstream region of GDH1 were amplified by PCR from CEN.PK113-5D genomic
12
DNA using primer pairs 23/24 and 25/26. The loxP-kanMX-loxP cassette was
13
amplified from plasmid pUG6 [59] as two overlapping fragments using primer pairs
14
29/30 (5´ part) and 31/32 (3´ part). By fusion PCR, the upstream region of GDH1 was
15
combined with the 5´ part of the kanMX cassette and the 3´ part of the kanMX
16
cassette with the downstream region of GDH1 and the resulting fragments used to
17
transform SCICK16. Transformation was performed using the standard lithium
18
acetate procedure [58] and transformants were selected using YPD plates containing
19
200 mg/l G418 (Formedium, Hunstanton, UK). Correct integration of the kanMX
20
cassette into the GDH1 locus was tested by PCR using primers 27/28. The kanMX
21
marker was subsequently excised by transient transformation with plasmid pSH47
22
containing the Cre recombinase encoding gene [59] leading to formation of strain
23
SCIGS03. Strain SCIGS06 carrying a genomic integration of genes ERG20 and
24
GDH2 under control of the TEF1 and PGK1 promoter, respectively, was obtained by
25
transforming strain SCIGS03 with the MreI/AscI fragment isolated from plasmid
All S. cerevisiae strains constructed in this work have a CEN.PK background with
- 22 -
1
pIGS05. Correct integration into the YORWΔ22 locus on chromosome XV [60] was
2
verified by PCR using primer pairs 33/1 and 5/34. Strain SCIGS12 carrying a
3
genomic integration of genes ERG20, GDH2, tHMG1 and upc2-1, was constructed by
4
co-transforming strain SCIGS03 with the MreI/AscI fragments isolated from plasmids
5
pIGS06 and pIGS10. Correct integration into the YORWΔ22 locus was verified by
6
PCR using primer pairs 33/1, 5/35, 19/15, 35/15, 16/36 and 13/34.
7
Kl URA3 was replaced in strains SCIGS06 and SCIGS12 with the kanMX marker. The
8
loxP-kanMX-loxP cassette was independently amplified from plasmid pUG6 using
9
primers pairs 37/38 for integration in strain SCIGS06 and 39/40 for integration in
10
SCIGS12 containing 71-74 bp primer tails complementary to the target integration
11
sites. Both strains were transformed with the respective PCR-amplified fragment.
12
Transformants were selected on YPD plates containing 200 mg/l G418. Kl URA3
13
replacement was initially tested by replica plating on synthetic complete (SC) medium
14
without uracil and YPD/G418 medium. The kanMX marker was subsequently
15
removed [59] leading to strains SCIGS22 and SCIGS23.
16
Strains SCIGS28, SCIGS29, SCIGS30, SCIGS31, SCIGS24 and SCIGS25 were
17
obtained transforming, respectively, strains CEN.PK113-5D, SCICK01, SCICK16,
18
SCIGS03, SCIGS22 and SCIGS23 with the high copy number plasmid pISP15 (Table
19
2) containing the URA3 gene and the genes SanSynopt and tHMG1 under control of the
20
strong constitutive promoters TEF1 and PGK1, respectively (Table 1).
21
22
Strain maintenance
23
performed in cryovials at -80ºC [61]. Working stocks were maintained on YPD agar
24
plates containing 10 g/l yeast extract, 20 g/l casein peptone, 20 g/l glucose and 20 g/l
25
agar. Plasmid carrying strains were maintained on synthetic dextrose medium agar
Long term storage of yeast suspensions containing 25% (vol/vol) sterile glycerol was
- 23 -
1
plates lacking uracil containing 6.9 g/l yeast nitrogen base without amino acids
2
(Formedium), 0.77 g/l complete supplement mixture without uracil (Foremedium) 20
3
g/l dextrose and 20 g/l agar.
4
5
Media and growth conditions
6
and had the following composition (per liter): (NH4)2SO4, 5 g; KH2PO4, 3 g;
7
MgSO4•7H2O, 0.50 g; Antifoam 289 (A204, Sigma–Aldrich), 0.05 ml; trace metals, 1
8
ml and vitamins, 1 ml. The trace metal solution consisted of the following (per liter):
9
EDTA (sodium salt), 15.0 g; ZnSO4•7H2O, 0.45 g; MnCl2•2H2O, 1 g; CoCl2•6H2O,
A mineral salts medium was used for batch cultivations as previously described [62]
10
0.3 g; CuSO4•5H2O, 0.3 g; Na2MoO4•2H2O, 0.4 g; CaCl2•2H2O, 0.45 g; FeSO4•7H2O,
11
0.3 g; H3BO3, 0.1 g and KI, 0.1 g. The pH of the trace metal solution was adjusted to
12
4.0 with 2 M NaOH prior to heat sterilization. The vitamin solution contained (per
13
liter): biotin, 0.05 g; p-amino benzoic acid, 0.2 g; nicotinic acid, 1 g; Ca-pantothenate,
14
1 g; pyridoxine-HCl, 1 g; thiamine-HCl, 1 g and myo-inositol, 25 g. The pH of the
15
vitamin solution was adjusted to 6.5 with 2 M NaOH. The vitamin solution was filter
16
sterilized and stored at 4°C. This medium was supplemented with 20 g/l glucose. The
17
feed composition used for continuous cultivation had the same composition as
18
described above, but the glucose concentration was 10 g/l. The medium used for
19
shake flask cultivation has the same composition as described above, but the
20
(NH4)2SO4 concentration was increased to 7.5 g/l, and the KH2PO4 to 14.4 g/l; the
21
glucose concentration was 20 g/l; the pH was adjusted to 6.5 prior autoclaving.
22
23
Inoculum preparation and pre-culture
24
flask containing 100 ml mineral salts medium. The seed culture was grown at 30ºC in
A single colony from an SC-ura agar plate was selected to inoculate a 500 ml shake
- 24 -
1
an orbital shaker at 100 rpm to late-exponential phase and used to inoculate the
2
fermenter to a final dry weight of 1 mg/l. All cultivations were performed in triplicate.
3
4
Chemostat operation
5
vessels (DasGip, Jülich, Germany) with a working volume of 0.3 l. The temperature
6
was monitored using a platinum RTD temperature sensor and kept at 30ºC using a
7
BioBlock integrated heating and cooling thermo well. Agitation was maintained at
8
600 rpm using an overhead drive stirrer with one Rushton impeller. The air flow rate
9
was kept at 1 vvm by a mass flow controller (DasGip). The pH was maintained
Aerobic, carbon limited chemostat cultivations were performed in 1.0 l stirrer pro
10
constant at 5.0 by automatic addition of 2 M KOH. The fermenters were integrated in
11
a DasGip monitor and control system used to control all fermentation parameters,
12
temperature, agitation, pH, and gas flow. Dissolved oxygen was monitored using an
13
autoclavable polarographic oxygen electrode (Mettler Toledo, Columbus, OH) and
14
maintained above 30% saturation via regulating stirrer speed and gas flow rate.
15
Exhaust gas was cooled, dried and the gas composition was analyzed for real time
16
continuous determination of oxygen and carbon dioxide concentration by a DasGip
17
fed batch pro® gas analysis system with off gas analyzer GA4 based on zirconium
18
dioxide and two-beam infrared sensor. The integrated mass flow sensor allowed on-
19
line monitoring and calculation of oxygen transfer rate (OTR), carbon dioxide transfer
20
rate (CTR) and respiratory quotient (RQ). The chemostat bioreactor was initiated as
21
batch culture with 10 g/l glucose. Only after the residual ethanol produced was
22
completely consumed the feed was started and the fermentation run in a continuous
23
mode. Fermenters were operated at dilution rate 0.05 or 0.1 h-1. A two-phase product
24
partition chemostat was performed by co-feeding medium containing 10 g/l glucose
25
and the organic phase (Figure 2). To obtain a dilution rate of 0.1 h-1, the inlet medium
- 25 -
1
was fed at 27 ml/h and the organic phase at 3 ml/h. To obtain a dilution rate of 0.05 h-
2
1
3
constant inlet feed ratio of medium:organic phase of 9:1 (vol/vol). Dodecane (Sigma-
4
Aldrich, St. Luis, MO) was used as organic phase and filter sterilized prior addition.
5
The culture working volume of 0.3 l (0.27 l of medium + 0.03 l of dodecane) was kept
6
constant by automatic withdrawal of broth based on an electric level sensor
7
measurement. The set-up allowed maintaining the correct medium/organic phase ratio
8
inside the fermentor throughout the fermentation time. The correct ratio of 9:1 vol/vol
9
between the two phases was constantly monitored and differed by less than 2% in
10
samples taken directly from the culture and from the effluent line. Steady state was
11
reached after at least 5 residence times, defined by constant values of CTR, OTR and
12
biomass concentration (less than 5% deviation).
13
14
Cell mass determination
15
and dry cell weight and on-line with an optical density transmitter OD4 sensor
16
(DasGip) integrated in the fermenter system. The optical density at 600 nm was
17
determined using a Genesis20 spectrophotometer (Thermo Scientific, Madison, WI,
18
USA). The cell dry weight was measured by filtering known culture volumes through
19
pre-dried and pre-weighed 0.45-μm-pore size nitrocellulose filters (Sartorious Stedim
20
Biotech GmbH, Göttingen, Germany). The filters with the biomass were washed with
21
water, dried for 15 min in a microwave oven at 150 W, and weighed again. The
22
correlation factor between off-line and on-line parameters was determined.
23
24
Metabolite analysis
25
two-phase steady state chemostat cultures and centrifuged for 5 min at 5000 g. The
, medium was feed at 13.5 ml/h and the organic phase at 1.5 ml/h resulting in a
Cell growth during fermentation was monitored off-line by measuring optical density
Samples for analysis of extracellular metabolite concentrations were withdrawn from
- 26 -
1
organic layer was discarded and the cultivation broth was filtered through 0.45-μm-
2
pore size nylon filters (VWR international, Radnor, PA, USA) and stored at -20ºC
3
until further analysis. Glucose, glycerol acetate, succinate, and pyruvate were
4
quantified by HPLC (UltiMate® 3000 Nano, Dionex, Bannockburn, IL, USA) with an
5
Aminex HXP-87H ion-exchange column (Bio-Rad, Hercules, CA) maintained at 65ºC
6
and using 5 mM H2SO4 as mobile phase at a flow rate of 0.6 ml min-1. Glucose,
7
glycerol, and ethanol were measured with a refraction index detector (RI-101
8
Refractive Index Detector, Shodex®), and acetate, succinate, and pyruvate were
9
measured with a UV-visible light absorbance detector (UltiMate 3000 Variable
10
Wavelength Detector, Dionex).
11
12
Analysis of sesquiterpenes
13
modifications. Culture samples were centrifuged 15 min at 5000 g and the organic
14
layer was diluted with an equal volume of dodecane containing a defined amount of
15
α-humulene as internal standard. Samples were diluted in heptane and analyzed by
16
gas chromatography-mass spectrometry (Thermo Scientific) equipped with an SLB-5
17
ms capillary column (30 m, 0.25 mm i.d., 0.25 µm film thickness; Supelco,
18
Bellefonte, PA, USA). Full mass spectra were generated by scanning the m/z range
19
within 40-500 for metabolite identification. Sesquiterpene identification was carried
20
out comparing mass spectra and retention time with authentic standards,
21
concentrations were calculated using a correction factor determined for the internal
22
standard -humulene relative to -santalene and E,E-farnesol.
23
24
25
Competing interests
Sequiterpene production was determined as described previously [12] with minor
GS, SP, VS and JN declare they have no competing interests. MS and LD are
employees of Firmenich SA.
- 27 -
1
2
Authors' contributions
3
project. G.S. performed the experimental work. S.P. assisted the molecular biology
4
experiments. M.D. and L.D. assisted the GC/MS analysis of sesquiterpens. G.S.
5
analyzed the data and wrote the manuscript. All the authors discussed the results,
6
edited and approved the final manuscript.
7
8
Acknowledgements
9
Foundation and Chalmers Foundation.
10
11
J.N. and G.S. participated in the design of the study. J.N. and V.S. supervised the
This work has been financed by Firmenich SA, the Knut and Alice Wallenberg
References
1. Leonard E, Ajikumar PK, Thayer K, Xiao W, Mo JD, Tidor B,
12
Stephanopoulos G, Prather KLJ: Combining metabolic and protein
13
engineering of terpenoid biosynthetic pathway for overproduction and
14
selectivity control. PNAS 2010, 107:13654-13659.
15
2. Daviet L, Schalk M: Biotechnology in plant essential oil production:
16
progress and perspective in metabolic engineering of the terpene
17
pathway. Flavour Fragr J 2010, 25: 123-127.
18
19
20
21
22
23
24
25
3. Kirby J, Keasling JD: Biosynthesis of plant isoprenoids: perspectives for
microbial engineering. Annual Rev Plant Biol 2009, 60: 335-355.
4. Chang MC, Keasling JD: Production of isoprenoid pharmaceuticals by
engineered microbes. Nat Chem Biol 2006, 2: 674-681.
5. Khalil AS, Collins JJ: Synthetic biology: applications come of age. Nat Rev
Gen 2010, 11: 367-379.
6. Fowler ZL, Koffas M: Microbial biosynthesis of fine chemicals: an
emerging technology. The metabolic pathway engineering handbook: tools
- 28 -
1
and applications. Volume 1. Edited by Smolke CD. Boca Raton: CRC Press;
2
2010.
3
7. Fortman JL, Chhabra S, Mukhopadhyay A, Chou H, Lee TS, Steen E,
4
Keasling JD: Biofuel alternatives to ethanol: pumping the microbial well.
5
Cell 2008, 26: 375-381.
6
8. Maury J, Asadollahi MA, Møller K, Clark A, Nielsen J: Microbial isoprenoid
7
production: an example of green chemistry through metabolic
8
engineering. Adv Biochem Eng Biotechnol 2005, 100: 19-51
9
9. Ohto C, Muramatsu M, Obata S, Sakuradani E, Shimizu S: Production of
10
geranylgeraniol on overexpression of a prenyl diphosphate synthase
11
fusion gene in Saccharomyces cerevisiae. Appl Microbiol Biotechnol 2010,
12
87: 1327-1334.
13
14
15
16
10. Schalk M: Method for producing alpha-santalene. US Pat 2011/008836 A1.
2011
11. Keasling JD: Synthetic biology for synthetic chemistry. ACS Chem Biol
2008, 3: 64-67.
17
12. Scalcinati G, Knuf C, Partow S, Chen Y, Maury J, Schalk M, Daviet L, Nielsen
18
J, Siewers V: Dynamic control of gene expression in Saccharomyces
19
cerevisiae engineered for the production of plant sesquitepene α-santalene
20
in a fed-batch mode. Metab Eng, doi:10.1016/j.ymben.2012.01.007.
21
13. Takahashi S, Yeo Y, Greenhagen BT, McMullin T, Song L, Maurina-Brunker
22
J, Rosson R, Noel JP, Chappell J: Metabolic engineering of sesquiterpene
23
metabolism in yeast. Biotechnol Bioeng 2007, 97: 170-181.
24
14. Asadollahi MA, Maury J, Schalk M, Clark A, Nielsen J: Enhancement of
25
farnesyl diphosphate pool as direct precursor of sesquiterpenes through
- 29 -
1
metabolic engineering of the mevalonate pathway in Saccharomyces
2
cerevisiae. Biotechnol Bioeng 2010, 106: 86-96.
3
15. Asadollahi MA, Maury J, Møller K, Nielsen KF, Schalk M, Clark A, Nielsen
4
J: Production of plant sesquiterpenes in Saccharomyces cerevisiae: Effect
5
of ERG9 repression on sesquiterpene biosynthesis. Biotechnol Bioeng 2008,
6
99: 666-677.
7
16. Faulkner A, Chen X, Rush J, Horazdovsky B, Waechter CJ, Carman GM,
8
Sternweis PC: The LPP1 and DPP1 gene products account for most of the
9
isoprenoid phosphate phosphatase activities in Saccharomyces cerevisiae.
10
J Biol Chem 1999, 274: 14831-14837.
11
17. Hampton RY, Rine J: Regulated degradation of HMG-CoA reductase, an
12
integral membrane protein of the endoplasmic reticulum, in yeast. J Cell
13
Biol 1994, 125: 299-312.
14
18. Polakowski T, Stahl U, Lang C: Overexpression of a cytosolic
15
hydroxymethylglutaryl-CoA reductase leads to squalene accumulation in
16
yeast. Appl Microbiol Biotechnol 1998, 49: 66-71.
17
18
19. Jackson BE, Hart-Wells EA, Matsuda SP: Metabolic engineering to produce
sesquiterpenes in yeast. Org Lett 2003, 5: 1629-1632.
19
20. Ro DK, Paradise EM, Ouellet M, Fisher KJ, Newman KL, Ndungu JM, Ho
20
KA, Eachus RA, Ham TS, Kirby J, Chang MCY, Withers ST, Shiba Y,
21
Sarpong R, Keasling JD: Production of the anitmalarial drug precursor
22
artemisinic acid in engineered yeast. Nature 2006, 440: 940-943.
23
21. Kirby J, Romanini DW, Paradise EM, Keasling JD: Engineering triterpene
24
production in Saccharomyces cerevisiae - beta-amyrin synthase from
25
Artemisia annua. FEBS J. 2008, 275: 1852-1859.
- 30 -
1
22. Anderson MS, J G Yarger, Burck CL, Poulter CD: Farnesyl diphosphate
2
synthetase. Molecular cloning, sequence, and expression of an essential
3
gene from Saccharomyces cerevisiae. J Biol Chem 1989, 264: 19176-19184.
4
23. Wang K, Ohnuma S: Chain length determination mechanism of isoprenyl
5
diphosphate synthases and implications for molecular evolution. TIBS
6
1999, 24: 445-451.
7
24. Daum G, Lees ND, Bard M, Dickson R: Biochemistry, cell biology and
8
molecular biology of lipids of Saccharomyces cerevisiae. Yeast 1998, 14:
9
1471-1510.
10
25. Chambon C, Ladeveze V, Oulmouden A, Servouse M, Karst F: Isolation and
11
properties of yeast mutants affected in farnesyl diphosphate synthetase.
12
Curr Genet 1990, 18: 41-46.
13
26. Grabinska K, Palamarczyk G: Dolichol biosynthesis in the yeast
14
Saccharomyces cerevisiae: an insight into the regulatory role of farnesyl
15
diphosphate synthase. FEMS Yeast Res 2002, 2: 259-265.
16
27. Szkopinska A, Swiezewska E. Karst F: The regulation of activity of main
17
mevalonic acid pathway enzymes: farnesyl diphosphate synthase, 3-
18
hydroxy-3-metylglutaryl-CoA reductase, and squalene synthase in yeast
19
Saccharomyces cerevisiae. Biochem Biophys Res Commun 2000, 267: 473-
20
477.
21
28. Hou J, Scalcinati G, Oldiges M, Vemuri GN: Metabolic impact of increased
22
NADH availability in Saccharomyces cerevisiae. Appl Environ Microbiol
23
2010, 76: 851-859.
- 31 -
1
29. Nissen TL, Kielland-Brandt MC, Nielsen J, Villadsen J: Optimization of
2
ethanol production in Saccharomyces cerevisiae by metabolic engineering
3
of the ammonium assimilation. Metab Eng 2000, 2: 69-77.
4
30. dos Santos M, Thygesen G, Kötter P, Olsson L, Nielsen J: Aerobic
5
physiology of redox-engineered Saccharomyces cerevisiae strains modified
6
in the ammonium assimilation for increased NADPH availability. FEMS
7
Yeast Res 2003, 4: 59-68.
8
31. Asadollahi M, Maury J, Patil KR, Schalk M, Clark A, Nielsen J: Enhancing
9
sesquiterpene production in Saccharomyces cerevisiae through in silico
10
11
12
driven metabolic engineering. Metab Eng 2009, 11: 328-334.
32. Vik Å, Rine J: Upc2p and Ecm22p, dual regulators of sterol biosynthesis
in Saccharomyces cerevisiae. Mol Cell Biol 2001, 21: 6395-6405.
13
33. Lewis TL, Keesler AG, Fenner GP, Parks W: Pleiotropic mutations in
14
Saccharomyces cerevisiae affecting sterol uptake and metabolism. Yeast
15
1988, 4: 93-106.
16
34. Davies BSJ, Wang HS, Rine J: Dual activators of the sterol biosynthetic
17
pathway of Saccharomyces cerevisiae: similar activation/regulatory
18
domains but different response mechanisms. Mol Cell Biol 2005, 25: 7375-
19
7385.
20
35. Paradise EM, Kirby J, Chan R, Keasling JD: Redirection of flux through the
21
FPP branch-point in Saccharomyces cerevisiae by down-regulating
22
squalene synthase. Biotechnol Bioeng 2008, 100: 371-378.
23
24
36. Nielsen J, Villadsen J, Liden G: Bioreaction Engineering Principles. New
York: Kluwer Plenum; 2003.
- 32 -
1
37. Raghevendran V, Patil KR, Olsson L, Nielsen J: Hap4 is not essential for
2
activation of respiration at low specific growth rates in Saccharomyces
3
cerevisiae. J Biol Chem 2006, 281: 12308-12314.
4
38. Diderich JA, Schepper M, Van Hoek P, Luttik MA, Van Dijken JP, Pronk JT,
5
Klaassen P, Boelens HF, Teixeira-de-Mattos J, Van Dam K, Kruckeberg AL:
6
Glucose uptake kinetics and transcription of HXT genes in chemostat
7
cultures of Saccharomyces cerevisiae. J Biol Chem 1999, 274: 15350–15359
8
39. Stark D, von Stockar U: In situ product removal (ISPR) in whole cell
9
biotechnology during last twenty years. Adv Biochem Eng Biotechnol 2003,
10
11
12
13
14
80: 150-175.
40. Van Sonsbeek HM, Beeftink HH, Tramper J: Two-phase bioreactors.
Enzyme Microb Technol 1993, 15: 722-729.
41. Christianson DW: Unearthing the roots of the terpenome. Curr Opin Chem
Biol 2008, 12: 141–150.
15
42. Wang C, Yoon H.S, Jang JH, Chung RY, Kim YJ, Choi SE, Kim WS:
16
Metabolic engineering of Escherichia coli for α-farnesene production.
17
Metab Eng, 13: 648-655.
18
43. Farhi M, Marhevka E, Masci T, Marcos E, Eyal Y, Ovadis M, Abeliovich H,
19
Vainstein A: Harnessing yeast subcellular compartments for the
20
production of plant terpenoids. Metab Eng 2011, 13: 474-481.
21
44. Madsen KM, Udatha GD, Semba S, Otero JM, Koetter P, Nielsen J, Ebizuka
22
Y, Kushiro T, Panagiotou G: Linking genotype and phenotype of
23
Saccharomyces cerevisiae strains reveals metabolic engineering targets
24
and leads to triterpene hyper-producers. Plos One 2011, 6: e14763
- 33 -
1
45. DeJong JM, Liu Y, Bollon AP, Long RM, Jennewein S, Williams D, Croteau
2
RB: Genetic engineering of taxol biosynthetic genes in Saccharomyces
3
cerevisiae. Biotechnol Bioeng 2005, 93: 212-224.
4
5
6
7
46. Arnezeder C, Hample WA: Influence of growth rate on the accumulation of
ergosterol in yeast-cells. Biotechnol Lett 1990, 12: 277-282.
47. Carman GM, Wu WI: Lipid phosphate phosphatases from Saccharomyces
cerevisiae. Methods Enzymol 2007, 434: 305-315.
8
48. Chiang DY, Moses AM, Kellis M, Lander ES, Eisen MB: Phylogenetically
9
and spatially conserved word pairs associated with gene-expression
10
11
12
changes in yeast. Genome Biol 2003, 4: R43.
49. Woehrer W, Roehr M: Regulatory aspects of Bakers’ yeast metabolism in
aerobic fed-batch cultures. Biotechnol Bioeng 23: 567-581.
13
50. Verduyn C, Zomerdijk TPL, van Dijken JP, Scheffers A: Continuous
14
measurement of ethanol production by aerobic yeast suspensions with an
15
enzyme electrode. Appl Microbiol Biotechnol 1984, 19: 181-185.
16
51. van der Rest ME, Kamminga AH, Nakano A, Anraku Y, Poolman B, Konings
17
WN: The plasma membrane of Saccharomyces cerevisiae: structure,
18
function, and biogenesis. Microbiol Rev 1995, 59: 304-322.
19
52. Verduyn C, Postma E, Scheffers WA, van Dijken JP: Physiology of
20
Saccharomyces cerevisiae in anaerobic glucose-limited chemostat cultures.
21
J Gen Microbiol 1990, 136: 395-403.
22
53. Özcan S, Johnston M: Three different regulatory mechanisms enable yeast
23
hexose transporter (HXT) genes to be induced by different levels of
24
glucose. Mol Cell Biol 1995, 15: 1564-1572.
- 34 -
1
54. Basso TO, Dario MG, Tonso A, Stambuk BU, Gombert AK: Insufficient
2
uracil supply in fully aerobic chemostat cultures of Saccharomyces
3
cerevisiae leads to respiro-fermentative metabolism and double nutrient-
4
limitation. Biotechnol Lett 2010, 32: 973-997.
5
55. Rodriguez-Limas W, Tyo KEJ, Nielsen J, Ramirez T, Palomares LA,
6
Molecular and process design for rotavirus-like particle production in
7
Saccharomyces cerevisiae. Microb Cell Fact 2011, 10: 33.
8
56. Reid RJD, Sunjevaric I, Kedacche M, Rothstein R: Efficient PCR-based gene
9
disruption in Saccharomyces strains using intragenic primers. Yeast 2002,
10
19: 319-328.
11
57. Van Dijken J P, Bauer J, Brambilla L, Duboc P, Francois JM, Gancedo C,
12
Giuseppin ML, Heijnen JJ, Hoare M, Lange HC, Madden EA, Niederberger P,
13
Nielsen J, Parrou JL, Petit T, Porro D, Reuss M, N. van Riel N, Rizzi M,
14
Steensma HY, Verrips CT, Vindelov J, Pronk JT:. 2000. An interlaboratory
15
comparison of physiological and genetic properties of four Saccharomyces
16
cerevisiae strains. Enzyme Microb Technol 2000, 26: 706-714.
17
58. Gietz RD, Woods RA: Transformation of yeast by lithium acetate/single
18
stranded carrier DNA/polyethylene glycol method. Methods Enzymol 2002,
19
350: 87-96.
20
59. Güldener U, Heck S, Fiedler T, Beinhauer J, Hegemann JH: A new efficient
21
gene disruption cassette for repeated use in budding yeast. Nucleic Acids
22
Res 1996, 24: 2519-2524.
23
60. Flagfeldt DB, Siewers V, Huang L, Nielsen J: Characterization of
24
chromosomal integration sites for heterologous gene expression in
25
Saccharomyces cerevisiae. Yeast 2009, 26: 545-551.
- 35 -
1
2
61. Sherman F, Fink GR, Hicks JB: Methods in yeast genetics – Laboratory
Manual. New York: Cold Spring Harbor; 1986.
3
62. Verduyn V, Postma E, Scheffers W.A, Van Dijken JP: Effect of benzoic acid
4
on metabolic fluxes in yeasts: A continuous-culture study on the
5
regulation of respiration and alcoholic fermentation. Yeast 1992, 8: 501-
6
517.
7
8
Figures
9
10
11
Figure 1 - Genetic engineering approach for increasing α-santalene
production.
12
reductase, a codon optimized santalene synthase gene (SanSynopt) PTEF1 and PPGK1
13
promoters as well as TADH1 and TCYC1 terminator sequences. (B) Integrated cassettes,
14
rectangles containing arrows represent the promoters and their directionality,
15
pentagons the genes and empty squares the terminators. (C) Scheme of the engineered
16
mevalonate, prenyl phosphate and ammonium assimilation pathways and FPP branch
17
point; overexpressed and deleted genes are highlighted. Pathway intermediates: G6P:
18
glucose-6-phosphate, Acetyl-CoAcyt: cytosolic acetyl-CoA, HMG-CoA: 3-hydroxy-3-
19
methylglutaryl-CoA, MVA: mevalonate, MVA-P: phosphomevalonate, MVA-PP:
20
diphosphomevalonate, IPP: isopentenyl diphosphate, DMAPP: dimethylallyl
21
diphosphate, GPP: geranyl diphosphate, FPP: farnesyl diphosphate, FOH: farnesol.
22
Overexpressed genes are tHMG1 (encoding truncated HMG-CoA reductase), ERG20
23
(encoding FPP synthase), GDH2 (encoding NAD-dependent glutamate
24
dehydrogenase), and SanSynopt (encoding α-santalene synthase). Deleted genes are
25
GDH1 (encoding NADP-dependent glutamate dehydrogenase), LPP1 and DPP1
26
(both encoding lipid phosphate phosphatases). The promoter of the ERG9 gene
(A) Expression plasmid pISP15 containing tHMG1 encoding truncated HMG-CoA
- 36 -
1
(encoding squalene synthase) is replaced with PHXT1. Genes whose promoters contain
2
Upc2 binding sites are indicated with a grey arrow: ERG13 (encoding HMG-CoA
3
synthase), ERG12 (encoding mevalonate kinase), and ERG8 (encoding
4
phosphomevalonate kinase). Additional genes indicated are ERG10 (encoding
5
acetoacetyl-CoA thiolase), ERG19 (encoding diphosphomevalonate decarboxylase)
6
and IDI (encoding IPP isomerase).
7
8
9
10
Figure 2 - Set-up of the in situ product removal (ISPR) chemostat cultivation
process.
11
system feeding culture medium and organic solvent. The product is continuously
12
captured in the organic phase due to its high hydrophobicity. In an integrated
13
downstream step the two phases of the effluent are partitioned in a settler.
14
Subsequently, the product is recovered from the organic phase, which can then be
15
further recycled in the same process. The exhausted medium is discarded.
16
17
18
Figure 3 - Sesquiterpene productivity in a two-phase partitioned glucoselimited aerobic chemostat.
19
weight of α-santalene and farnesol are, respectively, 13.62 and 14.82 g Cmol-1). (A)
20
Strains SCIGS28 (tHMG1↑), SCIGS29 (+ PHXT1-ERG9, lpp1Δ), SCIGS30 (+ dpp1Δ),
21
SCIGS31 (+ gdh1Δ), SCIGS24 (+ ERG20↑, GDH2↑), SCIGS25 (+ upc2-1↑,
22
tHMG1↑) cultivated at dilution rate D=0.05 h-1. (B) Strains SCIGS28 (tHMG1↑),
23
SCIGS29 (+ PHXT1-ERG9; lpp1Δ), SCIGS30 (+lpp1Δ) cultivated at dilution rate
24
D=0.1 h-1. Error bars represent the standard deviation from three independent
25
cultivations.
A stirred tank reactor is operated in continuous cultivation mode as double phase
α-Santalene and farnesol production rate in Cmmol (g biomass)-1 h-1 (the C-molar
- 37 -
1
2
3
Figure 4 - Sesquiterpene yield in a two-phase partitioned glucose limited
aerobic chemostat.
4
(tHMG1↑), SCIGS29 (+ PHXT1-ERG9, lpp1Δ), SCIGS30 (+ dpp1Δ) were cultivated at
5
dilution rate D=0.05 h-1 and D=0.1 h-1. Strains SCIGS31 (+ gdh1Δ), SCIGS24
6
(+ERG20↑, GDH2↑), SCIGS25 (+ upc2-1↑, tHMG1↑) were cultivated at dilution rate
7
D=0.05 h-1. Error bars represent the standard deviation from three independent
8
cultivations.
α-Santalene and farnesol yield in Cmmol (Cmmol glucose)-1. Strains SCIGS28
9
10
11
Figure 5 - α-Santalene titer in a two-phase partitioned glucose limited aerobic
chemostat.
12
molar weight of α-santalene and farnesol are respectively 13.62 and 14.82 g Cmol-1).
13
Strains SCIGS28 (tHMG1↑), SCIGS29 (+ PHXT1-ERG9, lpp1Δ), SCIGS30 (+ dpp1Δ)
14
were cultivated at dilution rate D=0.05 h-1 and D=0.1 h-1. Strains SCIGS31 (+ gdh1Δ),
15
SCIGS24 (+ERG20↑, GDH2↑), SCIGS25 (+ upc2-1↑, tHMG1↑) were cultivated at
16
dilution rate D=0.05 h-1. Error bars represent the standard deviation from three
17
independent cultivations.
α-Santalene (bottom) and farnesol (top) titers express in mg l-1 and Cmmol l-1 (the C-
18
19
Tables
20
Table 1 - Plasmids used in this study
21
Plasmid
name
Plasmid description
Reference
pSP-GM2
URA3-based expression plasmid carrying a
bidirectional PTEF1-PPGK1 promoter
PTEF1-SanSyn, PPGK1-tHMG1
PTEF1-SanSynopt, PPGK1-tHMG1
PTEF1-ERG20
PTEF1-ERG20, AD1
PTEF1-ERG20, PPGK1-GDH2, AD1
PTEF1-ERG20, PPGK1-GDH2, AD1, AD2
[55]
pICK01
pISP15
pIGS01
pIGS02
PIGS03
pIGS04
- 38 -
[12]
this study
this study
this study
this study
this study
pIGS05
pIGS06
pIGS07
pIGS08
pIGS09
pIGS10
1
PTEF1-ERG20, PPGK1-GDH2, AD1, AD2,
KlURA3
PTEF1-ERG20 PPGK1-GDH2, AD1, 5´KlURA3
PTEF1-tHMG1
PTEF1-tHMG1, PPGK1-upc2-1
PTEF1-tHMG1, PPGK1-upc2-1, AD3
PTEF1-tHMG1, PPGK1-upc2-1, AD3, 3´KlURA3
- 39 -
this study
this study
this study
this study
this study
this study
1
Table 2 - List of S. cerevisiae strains used in this study
Strain
Genotype
Plasmid Reference
c
CEN.PK113-5D MATa MAL2-8 SUC2 ura3-52
none
P. Kötter, University
of Frankfurt, Germany
c
SCIGS28
MATa MAL2-8 SUC2 ura3-52
pISP15 this study
SCICK01
MATa MAL2-8c SUC2 ura3-52
none
[12]
lpp1Δ::loxP PERG9Δ::loxP-PHXT1
SCIGS29
MATa MAL2-8c SUC2 ura3-52
pISP15 this study
lpp1Δ::loxP PERG9Δ::loxP-PHXT1
SCICK16
MATa MAL2-8c SUC2 ura3-52
none
[12]
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1
SCIGS30
MATa MAL2-8c SUC2 ura3-52
pISP15 this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1
SCIGS03
MATa MAL2-8c SUC2 ura3-52
none
this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP
SCIGS31
MATa MAL2-8c SUC2 ura3-52
pISP15 this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP
SCIGS06
MATa MAL2-8c SUC2 ura3-52
none
this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP
PTEF1-ERG20 PPGK1-GDH2
KlURA3
SCIGS22
MATa MAL2-8c SUC2 ura3-52
none
this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP
PTEF1-ERG20 PPGK1-GDH2
SCIGS24
MATa MAL2-8c SUC2 ura3-52
pISP15 this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP
PTEF1-ERG20 PPGK1-GDH2
SCIGS23
MATa MAL2-8c SUC2 ura3-52
none
this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP
PTEF1-ERG20 PPGK1-GDH2 PTEF1tHMG1 PPGK1-upc2-1
SCIGS12
MATa MAL2-8c SUC2 ura3-52
none
this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP
PTEF1-ERG20 PPGK1-GDH2 PTEF1tHMG1 PPGK1-upc2-1 KlURA3
SCIGS25
MATa MAL2-8c SUC2 ura3-52
pISP15 this study
lpp1Δ::loxP dpp1Δ::loxP
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP
PTEF1-ERG20 PPGK1-GDH2 PTEF1- 40 -
tHMG1 PPGK1-upc2-1
- 41 -
Table 3 - Physiological parameters measured during double-phase chemostat cultures of strains SCIGS28, SCIGS29, SCIGS30,
SCIGS31, SCIGS24 and SCIGS25.
μmax, specific growth rate (h-1); D, dilution rate (h-1); Ysx, biomass yield (g biomass (g substrate)-1); specific consumption rates of glucose(rs),
and oxygen (rO2) (mmol (g biomass)-1h-1); specific production rates of carbon dioxide (rCO2), ethanol (retoh),and acetate (racet) (mmol (g biomass)1 1
h- ). RQ, respiratory quotient rCO2/rO2; Cs, residual glucose concentration (mM); Cbalance, carbon recovery (%) Values represent the mean ±S.D.
of three independent cultivations.
Strain
D
Yxs
rs
rCO2
rO2
retoh
(h-1)
(g g-1)
0.051 ±0.002
0.50 ±0.01
0.57 ±0.01
1.12 ±0.08
1.12 ±0.06
0
0.10 ±0.01
0.50 ±0.01
1.11 ±0.03
2.67 ±0.15
2.49 ±0.07
0.050 ±0.003
0.29 ±0.01
0.97±0.04
3.17 ±0.04
0.10 ±0.01
0.28 ±0.01
1.95 ±0.05
0.051 ±0.001
0.28 ±0.02
0.10 ±0.02
SCIGS31
racet
RQ
Cs
Cbalance
(rCO2/rO2)
(mM)
(%)
0
1.00 ±0.02
0.18 ±0.02
100.3 ±2.1
0
0
1.07 ±0.05
0.16 ±0.01
101.9 ±1.1
2.95 ±0.15
0.082 ±0.001
0.024 ±0.001
1.07 ±0.07
1.16 ±0.04
96.9 ±3.2
6.40 ±0.18
5.51 ±0.20
0.128 ±0.004
0.039 ±0.008
1.16 ±0.22
0.39 ±0.01
95.0 ±0.9
1.09 ±0.05
3.60 ±0.09
3.41 ±0.22
0.099 ±0.015
0.024 ±0.003
1.05 ±0.09
1.15 ±0.04
94.7 ±0.4
0.25 ±0.01
2.26 ±0.09
7.70 ±0.19
6.37 ±0.29
0.161 ±0.007
0.032 ±0.005
1.21 ±0.14
0.40 ±0.01
93.5 ±1.2
0.051 ±0.001
0.33 ±0.01
0.86 ±0.01
2.31 ±0.08
1.42 ±0.23
0.501 ±0.077
0.047 ±0.007
1.62 ±0.11
30.67 ±0.78
105.4 ±4.9
SCIGS24
0.051 ±0.001
0.24 ±0.01
1.16±0.03
3.85 ±0.05
3.21 ±0.07
0.185 ±0.003
0.020 ±0.008
1.20 ±0.02
2.53 ±0.09
93.7 ±5.3
SCIGS25
0.048 ±0.003
0.21 ±0.01
1.26 ±0.02
4.41 ±0.04
3.82 ±0.04
0.195 ±0.005
0.027 ±0.007
1.16 ±0.03
2.91 ±0.14
93.8 ±2.9
(mmol g biomass-1 h-1)
SCIGS28
SCIGS29
SCIGS30
- 42 -
Additional files
Additional file 1 – Primers used in this study
Restriction sites are indicated in bold face, overlapping nucleotides are underlined,
and the modified codon is indicated in italic.
Additional file 2 –Codon optimized santalene synthase nucleotide sequence
Additional file 3 – Maps of plasmids constructed in this study
(A) pISP15, (B) pIGS01, (C) pIGS02, (D) pIGS03, (E) pIGS04, (F) pIGS05, (G)
pIGS06, (H) pIGS07, (I) pIGS08, (L) pIGS09, (M) pIGS10.
- 43 -
A
B
Expression plasmid
Integrated Cassettes
upc2-1
tHMG1
PPGK1 PTEF1
TADH1
ERG20
GDH2
NH4+
OUT
IN
C
Ammonium
assimilation
Glucose
Glucose
NH4+
+
2-Ketoglutarate
ERG10
Acetoacetyl-CoA
GDH1
Glutamate
GDH2
ERG13
MVA-P
ERG8
MVA-PP
ERG20
OPP
FPP
- 44 -
OH
FOH
GPP
CoA
Acetyl-CoA Cyt
DPP1
FPP branch point
OH
MVA
COOH
ERG12
LPP1
ERG20
HO
ERG19 IPP IDI1 DMAPP
Prenyl diphosphate pathway
tHMG1
tHMG1
HMG-CoA
Mevalonate pathway
PDH Bypass
Glycolysis
G6P
TCYC1
PPGK1 PTEF1
TADH1
2μ
TCYC1
PHXT1
ERG9
Squalene
SanSyn
α-Santalene
Ergosterol
Samples
Feed IN
Product recovery
Solvent
regeneration
Feed OUT
Reactor
Organic Solvent
Substrate
Product
Solvent recycle
- 45 -
Discharge
0,05
Santalene
Farnesol
Santalene and Farnesol
(Cmmol/gDCW h)
0,04
0,03
0,02
0,01
0
SCIGS28
SCIGS29
SCIGS30
SCIGS31
Strain
0,05
Santalene
Farnesol
Santalene and Farnesol
(Cmmol/gDCW h)
0,04
0,03
0,02
0,01
0,00
SCIGS28
SCIGS29
Strain
- 46 -
SCIGS30
SCIGS24
SCIGS25
0,008
Farnesol
Sanatalene
yield (Cmmol/Cmmol)
Sesquiterpene
0,007
Santalene
0,006
0,005
0,004
0,003
0,002
0,001
0,000
Dilution rate
Strain
0,05 0,1
SCIGS28
0,05 0,1
SCIGS29
0,05 0,1
SCIGS30
- 47 -
0,05
SCIGS31
0,05
SCIGS24
0,05
SCIGS25
25
20
1,6
0,05
0,1
0,8
10
0,4
5
0
25
Sanatalene (mg/l)
20
Farnesol (Cmmol/l)
15
0
0,05
0,1
1,6
1,2
15
0,8
10
0,4
5
0
0
SCIGS28
SCIGS29
SCIGS30
SCIGS31
Strain
- 48 -
SCIGS24
SCIGS25
Santalene (Cmmol/l)
Farnesol (mg/l)
1,2
PAPER III
Optimization of fed batch process for production of a
sesquiterpene biofuel-like precursor α-santalene by
Saccharomyces cerevisiae.
Scalcinati G and Nielsen J.
Submitted.
64
1
Optimization of fed batch process for production of sesquiterpene biofuel-like precursor
2
α-santalene by Saccharomyces cerevisiae
3
4
Gionata Scalcinati1 and Jens Nielsen1*
5
1
6
Chalmers University of Technology, Kemivägen 10, SE-412 96 Gothenburg, Sweden.
Systems & Synthetic Biology, Department of Chemical and Biological Engineering,
7
8
*
9
Professor Jens Nielsen
10
Tel: +46 31 772 38 04
11
Fax: +46 31 772 38 01
12
E-mail: [email protected]
Corresponding author
13
1
1
Abstract
2
The development of an efficient microbial production as sustainable technology has an
3
important impact for our society. Advanced biofuel are an emerging target product for
4
develop of novel cell factories. Various sequiterpene compounds have been proposed based
5
on their propriety as suitable biofuel precursor alternative. In the present study, we used a
6
previous engineered S. cerevisiae sesquiterpene producing strains to develop an efficient and
7
scalable fed-batch production process. By using these strains as platform host we perform a
8
process engineer and identified some of the process related bottlenecks. A respiratory
9
metabolic control was applied to a rationally designed exponential feed policy. This strategy
10
allowed to control feed delivery and successfully optimize the productivity, representing a
11
valuable tool for implement the production process. Additionally we investigate the effect of
12
ethanol as alternative carbon source to increase the precursor’s pool. Detailed process analysis
13
identified the excretion of cyclic sequiterpene as the major process limiting step.
14
15
Keyword: Fed-batch fermentation; Isoprenoids; Biofuel; Saccharomyces cerevisiae
16
2
1
Introduction
2
The demand for microbial production of chemicals as alternative to petrochemical based
3
synthesis is increasing due to economical, environmental and geopolitical factors
4
(Dellomonaco et al., 2010; Stephanopoulos et al., 2007). Microbial productions are gaining
5
popularity especially for the biosynthesis of added values compounds (Hong et al., 2012; Kim
6
et al., 2012). Recently, the role of isoprenoids as biomaterial resource has been rediscovered
7
leading to a renewed interest in this class of molecules (Bohlmann et al., 2008). Isoprenoids
8
represent an example of industrially relevant product targeted for bio-production, particularly
9
C15-branched sesquiterpenes beside their application has cosmetics, fragrance, flavouring,
10
nutraceutical and pharmaceutical agents (Leonard et al., 2010; Daviet et al., 2010), are
11
receiving attention as precursors of new generation biofuel for diesel and jet fuels (Peralta-
12
Yahya et al., 2011; Zhang et al., 2011; Rude et al., 2009; Lee et al., 2008). In the last decade,
13
the bio-based production of isoprenoids compound that can complement traditional producing
14
methods has progress dramatically (Westfall et al., 2011; Ro et al., 2006). Advances in
15
metabolic engineering enabled the availability of new technology tools and methods for
16
optimizing metabolic pathways leading to specific strategy for cell bioengineering (Chandran
17
et al., 2011; Keasling et al., 2010). Specifically the application of systems and synthetic
18
biology accelerates the process to achieve the ultimate goal of an efficient “microrefinery”
19
(Keasling et al., 2012; Siddiqui et al., 2012; Nielsen et al., 2011). In yeast sesquiterpenes (C15)
20
are synthesized from cytosolic acetyl-CoA though the mevalonate (MVA) pathway in sub-
21
sequential condensation of three isoprene functional unit isopentenyl diphosphate (C5)
22
resulting into the universal sesquiterpene building block farnesyl diphosphate (Fig. 1B)
23
(Maury et al., 2005).
24
Conversions of the liner FPP into cyclic derivative involve limited numbers of mechanisms
25
and result in the production of diverse classes of sesquiterpenes; example of FPP cyclization
3
1
products via nerolidyl diphosphate intermediate (NPP) are the structurally related
2
sesquiterpene classes: cedranes, santalanes, bisabolanes and bregamotanes (McCaskill et al.,
3
1997). The portfolio of fuel candidate compound has been greatly expanded lately. Recent
4
researches highlight branched and cyclic sesquiterpene hydrocarbons (e.g. bisabolanes and
5
farnesanes) as potential jet fuel precursor’s based on their physicochemical proprieties
6
(Peralta-Yahya et al., 2011; Renninger et al., 2008). Previously, we engineered S. cerevisiae
7
through a multistep metabolic engineering approach for the production of cyclic hydrocarbon
8
sesquiterpene α-santelene with suitable characteristics as fuel like precursor (Scalcinati et al.,
9
2012a; Scalcianti et al., 2012b).
10
Here, we used our yeast sesquiterpene producing platform to develop an efficient
11
fermentation process. Bioreactor operations have strong impact on design an efficient
12
bioprocess and influence in a significant way the production performances. Economically
13
feasible bio-production requires the development of a production process that allow to
14
achieve high titer, yields and productivities of targets compounds. Fed-batch cultivation is by
15
far the most employed fermentation process for industrial production (Nielsen et al., 2003).
16
Among the advantage of using fed-batch process for large scale process are the high final tiers
17
achievable and the minimization of overflow metabolism. During fed batch operation the feed
18
rate represent key parameter in controlling the overall process performances. Typical fed
19
batch configurations of high values products are based on a first phase were the feed is kept
20
exponential and a second phase when high cell concentration is reached with constant feed
21
rate to avoid potential limitation, however this often result in suboptimal conditions affecting
22
productivity and viability of the cell (Pham et al., 1998).
23
Operational strategies that allow optimal feeding policy are required to overcome problems
24
related to the detrimental effect due to the over/under feeding (Woehrer et al., 1981). Advance
25
in fermentation technology produce a multitude of strategy focused on properly control the
4
1
fed batch process (Lee et al., 1999). Ideal glucose based fed batch proceed maintaining a
2
sugar concentration below the critical values preventing the consequent Crabtree effect, if this
3
value is exceeded yeast metabolism rapidly switch to fermentative mode and the fermentation
4
products ethanol, acetate formation set in producing a reduction in the biomass yield.
5
In our previous study we elucidate the specific growth rate dependence of sesquiterpene
6
productivity of our engineered overproducing mutants (Scalcinati et al., 2012b). An optimal
7
process design would then require simultaneous formation of biomass and target compound
8
and would be favored without loss of carbon to fermentative over-flow metabolites.
9
Here, we develop a biomass-coupled production process applying a RQ control to a limited
10
glucose exponential fed batch; we use instant RQ measurement as indicator of the metabolic
11
state to fine tuning the feed profile to the cellular demand in order to maintain fully
12
respiratory state and achieve an optimal feed strategy. This strategy was combined with a
13
consolidated double-phase partitioning set-up resulting in an integrated fermentation recovery
14
process (Figure 1A).
15
Additionally, the effect of ethanol as alternative carbon source on sesquiterpene production
16
was investigated. The C2 carbon ethanol represents an attractive carbon source for secondary
17
metabolite production. However because the raw material is often the dominating operative
18
cost of added values chemicals industrial bio-production (Otero et al., 2007) a fully ethanol
19
based process would probably not be commercially viable. Yeast S. cerevisiae has the ability
20
to co consume glucose-ethanol under fully respiratory conditions (Geurts et al., 1980), if the
21
ratio between the two C-source is maintained below certain limits (0.57:0.43 Cmol Cmol-1)
22
glucose would be employed from the cell mainly for biosynthesis whereas all the cytosolyc
23
acetyl-CoA would be derived from ethanol that would then be used in the TCA cycle (de
24
Jong-Gubbels et al., 1995; van Gulik et al., 1995). We decide therefore to investigate the
25
effect of a mixed glucose/ethanol feed applied to the designed fed-batch process.
5
1
In this study, we present progress in production of sesquiterpene α-santalene, development in
2
fermentation method result in the creation of an efficient production system for α-santalene
3
capable of an overall yield, productivity and titers of 0.0043 Cmmol (Cmmol substrate)-1,
4
0.051 Cmmol (gDCW h)-1 and 14.2 Cmmol l-1.
5
6
Materials and methods
7
Strains and maintenance
8
S. cerevisiae strains used in this study have been previously engineered for α-sanatlene
9
production. They are SCIGS24 (MATa MAL2-8c SUC2 ura3-52 lpp1Δ::loxP dpp1Δ::loxP
10
PERG9Δ::loxP-PHXT1 gdh1Δ::loxP PTEF1-ERG20 PPGK1-GDH2) and SCIGS25 (MATa MAL2-8c
11
SUC2 ura3-52 lpp1Δ::loxP dpp1Δ::loxP PERG9Δ::loxP-PHXT1 gdh1Δ::loxP PTEF1-ERG20
12
PPGK1-GDH2 PTEF1-tHMG1 PPGK1-upc2-1 ) both strains carry the plasmid pISP15 (2μ URA3)
13
expressing a copy of the truncate HMG1 (tHMG1) and codon optimized Sansyn (SanSynOpt)
14
under control of the PGK1 and TEF1 promoter, respectively (Scalcinati et al., 2012b). Strains
15
were maintained on synthetic dextrose (SD) plate containing 6.9 g l-1 yeast nitrogen base w/o
16
amino acids (Formedium, Hunstanton, UK), 0.77 g l-1 complete supplement mixture (CSM)
17
w/o uracil (MP Biomedicals, Solon, OH, USA), 20 g l-1 glucose, and 20 g l-1 agar. Stock
18
cultures were kept in 30% (v/v) glycerol at -80ºC for long term storage (Sherman et al., 1986).
19
Media and growth conditions
20
A previously described (Verduyn et al., 1992) mineral salts medium was used consisting of
21
the following (per liter): (NH4)2SO4, 5 g; KH2PO4, 3 g; MgSO4•7H2O, 0.50 g; Antifoam 289
22
(A-5551, Sigma–Aldrich), 0.050 ml; trace metals, 1 ml and vitamins, 1 ml. The trace metal
23
solution consisted of the following (per liter): EDTA (sodium salt), 15.0 g; ZnSO4•7H2O, 0.45
24
g; MnCl2•2H2O, 1 g; CoCl2•6H2O, 0.3 g; CuSO4•5H2O, 0.3 g; Na2MoO4•2H2O, 0.4 g;
25
CaCl2•2H2O, 0.45 g; FeSO4•7H2O, 0.3 g; H3BO3, 0.1 g and KI, 0.10 g. The pH of the trace
6
1
metal solution was adjusted to 4.0 with 2 M NaOH prior to heat sterilization. The vitamin
2
solution contained (per liter): biotin, 0.05 g; p-amino benzoic acid, 0.2 g; nicotinic acid, 1 g;
3
Ca-pantothenate, 1 g; pyridoxine-HCl, 1 g; thiamine-HCl, 1 g and myo-inositol, 25 g. The pH
4
of the vitamin solution was adjusted to 6.5 with 2 M NaOH. The vitamin solution was filter
5
sterilized and stored at 4°C. This medium was supplemented with 30 g l-1 glucose. The feed
6
media had the same composition as described above, but the (NH4)2SO4; KH2PO4;
7
MgSO4•7H2O, vitamin solution, and trace metal solution, concentration were increased 10
8
times. The carbon source used was 100% glucose alone or a mixture of 57% Cmol glucose
9
and 43% Cmol ethanol. For the cultivations on glucose the target concentration was 6.66
10
Cmol l-1 (200 g l-1); for mixed carbon source cultivation was 3.80 Cmol l-1 (114.0 g l-1)
11
glucose and 2.86 Cmol l-1 (65.98 g l-1) ethanol, yielding a target final carbon concentration of
12
6.66 Cmol l-1. Both the sugar solutions were added by sterile filtration using a cellulose
13
acetate filter (0.20 m pore size Minisart -Plus Satorius AG).
14
Fed-batch operation
15
The aerobic fed-batch process was performed in 2.5 l Applikon vessels (Applikon, Schiedam,
16
The Netherlands) with an initial working volume of 1.0 l. Agitation at 800 rpm was
17
maintained using an integrated stirrer (DasGip, Jülich, Germany) and the temperature was
18
monitored using platinum RTD temperature sensor and kept at 30°C. The rate of aeration was
19
set to 1 vvm by mass flow controller. The pH of the medium was maintained at 5.0 by
20
automatic addition of 2 N KOH during the batch phase and 6 N KOH during the feed phase.
21
The temperature, agitation, gassing, pH and composition of the off-gas were monitored and
22
controlled using the integated DasGip monitoring and control system. Dissolved oxygen
23
concentration was monitored with an autoclavable polarographic oxygen electrode (Mettler
24
Toledo, Columbus, OH, USA) and kept above 30% via stirrer speed and gas flow rate using
25
the DasGip control system. The effluent gas from the fermentation was cooled, dried and
7
1
analyzed for real-time determination of oxygen and CO2 concentration by DasGip fedbatch
2
pro® gas analysis systems with the off gas analyzer GA4 based on zirconium dioxide and
3
two-beam infrared sensor. The integrated mass flow sensor allowed on-line monitoring and
4
calculation of oxygen transfer rate (OTR) (mmol h-1), carbon dioxide transfer rate (CTR)
5
(mmol h-1) and respiration quotient (RQ) calculate dividing CTR by OTR. The seed cultures
6
for the cultivations were grown at 30˚C in 500-ml shake flasks containing 100 ml of culture
7
with agitation in an orbital shaker at 100 rpm. Pre-cultures were used to inoculate the
8
fermentors to a final dry weight of 1 mg l-1. All cultivations were performed in triplicate. The
9
fed-batch cultures were initiated as batch cultures using 30 g l-1 glucose. Feeding with fresh
10
medium commenced only after residual ethanol produced from the glucose consumption
11
phase was completely depleted. An exponential feed rate ν (t) (liter h-1) was calculated as
12
previously described (Scalcinati et al., 2012a). The feed was initially designed to increase
13
exponentially with a specific feed rate of 0.1 h-1. Correct feed rate addition was obtained using
14
a modified single loop, set-point control method programming the fermenter fb-pro software
15
(DasGip) and controlled using the DasGip control system (Fig. 1A). Feed rate was
16
additionally controlled based on the respiratory culture response monitoring RQ values as
17
indirect measure of the cell metabolic state. On line precise measurement of the exit gas
18
composition allows to adjust the feed rate through a proportional integral (PI) controller
19
maintaining the RQ at the desired set point of 1.0 during feed with glucose alone and 0.82 for
20
glucose ethanol (0.57: 0.43 Cmol Cmol-1) mixed feed. An organic layer of dodecane (Sigma-
21
Aldrich St. Luis, MO) was added aseptically to a final volume of 10% (v/v) immediately
22
before starting the feed.
23
Analytical methods
24
Cell growth during fermentation was monitored off-line by measuring optical density and dry
25
cell weight and on-line with an optical density transmitter OD4 sensor (DasGip, Jülich,
8
1
Germany) integrated in the fermenter system. The optical density at 600 nm was determined
2
using a Genesis20 spectrophotometer (Thermo scientific, Madison WI, USA). The cell dry
3
weight was measured by filtering known volumes of the cultures through pre-dried and pre-
4
weighed 0.45-μm-pores size nitrocellulose filters (Supor-450 membrane filters; PALL Life
5
Sciences Ann Abor, MI). The filters with the biomass were washed with water, dried for 15
6
min in a microwave oven at 150 W, and weighed again.
7
Concentrations of glucose, glycerol, ethanol, acetate, succinate, and pyruvate were analysed
8
by an isocratic high-performance liquid chromatograph (UltiMate® 3000 Nano, Dionex,
9
Bannockburn IL, USA) with an Aminex HXP-87H ion-exchange column (Bio-Rad, Hercules,
10
CA) at 65˚C using 5 mM H2SO4 as mobile phase at a flow rate of 0.6 ml min-1. Glucose,
11
glycerol, and ethanol were measured with a refraction index detector (RI-101 Refractive
12
Index Detector, Shodex®), and acetate, succinate, and pyruvate were measured with a UV-
13
visible light absorbance detector (UltiMate 3000 Variable Wavelength Detector, Dionex)
14
Analysis of sesquiterpenes
15
Sequiterpene production during the course of fermentation was determined as described
16
previously (Scalcinati et al., 2012b) with minor modifications. For determine extracellular
17
sesquiterpene concentration culture samples were centrifuged 5 min at 5000 g the organic
18
layer collected, diluted with an equal volume of 2-ethylacetate (Sigma, St Luis, MO, USA)
19
containing defined amount of α-humulene as internal standard and subsequently analyzed by
20
gas chromatography-mass spectrometry (GC-MS). For determine the total intracellular
21
sesquiterpene concentration a repeated extractive process was used. A known volume of
22
cultures was harvested by centrifuging at 5000 rpm for 20 min. The cell pellet was washed
23
twice with distilled water and the cell suspension was centrifuged for another 10 min at 5000
24
rpm. The cell pellet was re-suspended in 500 μl of 2-ethylacetate and mixed with 250 mg of
25
acid washed glass beads (Sigma, St Luis, MO, USA) in a glass GC vial PTFE lined screw cap,
9
1
heated and mixed in a vortex termomixer at 60 ºC for 20 min and then centrifuged 15 min at
2
5000 rpm. The 2-ethylacetate layer was transferred to a clean GC vials and the pellet
3
undergoes to a new extractive cycle. A total of five cycles of extraction were necessary to
4
completely recover all the intracellular sesquiterpenes. The pooled extracts were diluted with
5
an equal volume of 2-ethylacetate containing defined amount of α-humulene as internal
6
standard and subsequently analyzed by GC-MS. The GC/MS sesquiterpene quantitative
7
determination was performed with a DSQ II single quadrupole mass spectrometer (Thermo
8
Scientific, Waltham, MA) equipped with a SLB-5 ms capillary column (15 m, 0.25 mm i.d.,
9
0.25 µm film thickness; Supelco, Bellefonte, PA, USA) using helium as carrier gas at a flow
10
rate of 1.2 ml min-1. The initial oven temperature was 80°C and the injector temperature was
11
250°C. The oven temperature was increased to 120°C at a rate of 10°C min-1 and
12
subsequently increased to 160°C at a rate of 3°C min-1. The oven temperature was finally
13
increased to 270°C at a rate of 10°C min-1 and held for 5 min at this temperature. Full mass
14
spectra were generated by scanning m/z range within 40-500. Sesquiterpene identification
15
was performed comparing mass spectra and retention time with authentic standards,
16
concentrations were calculated using a correction factor determined for the internal standard
17
-humulene relative to -santalene, trans-α-bergamotene, E,E-farnesene and E,E-farnesol.
18
19
Results
20
Performance of RQ controlled feeding strategy on double phase fed batch cultivation
21
Large scale commercial productions of microbial products are mainly realized using fed-batch
22
process (Nielsen et al., 2003). Recently we demonstrate the feasibility of sesquiterpene α-
23
santalene production in S. cerevisiae. (Scalcinati et al., 2012a). In order to increase
24
sesquiterpene yield, yeast strains were previously engineered through a multistep metabolic
25
engineered strategy, overexpressing key regulatory gene of the MVA pathway farnesyl
10
1
diphosphate synthase (ERG20 ) and a constitutive active form of 3-hydroxy-3-methyl-
2
glutaryl-CoA reductase (tHMG1 ); down-regulating precursor consuming reaction squalene
3
synthase ( ERG9); deleting gene catalyzing by-product formation (ΔLPP1, ΔDPP1);
4
increasing cofactors availability (ΔGDH1, GDH2); all this modification were integrated in
5
the genome to ensure genetic stability and lead to strain SCIGS24. Additional the
6
overexpression of a mutated global transcription factor (upc2-1) resulted in strain SCIGS25
7
(Scalcinati et al 2012b). Here, we undertook our previously engineered S. cerevisiae
8
sesquiterpene producing platform to design an efficient ISPR aerobic fed-batch production
9
process (Fig. 1A). The cultivation process was initiated as batch with 1 Cmol l-1 (30 g l-1)
10
glucose as carbon source, production phase was started feeding concentrate substrate after the
11
ethanol produced during the first glucose batch phase was completely consumed and no other
12
carbon sources were left in the media. Fed-batch was carried out as two phase cultivation
13
mode applying an organic overlay that allow in situ product recovery (Fig. 1A). Feed rate was
14
initially designed with exponential policy chosen so that the volumetric rate of biomass
15
production was constant and equal to a specific growth rate of 0.1 h-1. Additional RQ control
16
strategy was applied to the traditional exponential feed profile in order to maintain a fully
17
respiratory metabolism during the entire feed process maximizing the biomass formation and
18
avoiding the fermentation byproduct accumulation. Complete oxidation of glucose to carbon
19
dioxide and water through respiratory metabolism would lead to a respiratory coefficient RQ
20
equal to 1. On-line measurement of CTR and OTR allow instantaneous calculation of the RQ,
21
allowing to control the feed addition regulating the RQ close to the selected set point and
22
below the critical level that trigger the overflow metabolism (RQ 1). In an initial set of
23
experiment fed-batch was performed cultivating strains SCIGS24 and SCIGS25 using glucose
24
with a concentration of 7.4 Cmol l-1 (222 g l-1) as sole carbon source in the feed media. The
25
results of the respiratory quotient control on the feed profile are reported in figure 2. Both
11
1
strains display similar characteristics, within the first 15-16 hours of feed the RQ value
2
continue to increase, when the set-point value of 1 was reached the controller respond
3
decreasing the feed addition and the cultivation continued with a decrease exponential rate
4
lower than the theoretically designed with the RQ values oscillate between 1 and 1.2 (Fig. 2).
5
As result the metabolism was maintained fully respirative during the entire feed time and no
6
accumulation of glucose or fermentative products were detected in both strains.
7
Extractive fermentation was realized using a double phase partitioning set-up, so the product
8
is constantly removed from the cell and trapped into the solvent phase. Organic layer was
9
added before initiating the feed addition corresponding to the start of the production phase. In
10
order to monitor the sesquiterpene production performance during cultivation time, product
11
accumulation in the organic layer was analyzed at regular intervals. Four different
12
sesquiterpene products were detected, with α-santalene as major product followed by E,E-
13
farnesol, and minor amount of trans-α-bergamotene and E,E-farnesene (Fig. 3). Yield and
14
total sesquiterpene concentration were slightly higher in strain SCIGS24 than in strain
15
SCIGS25 (table I). Target compound α-santalene accumulate on strain SCIGS24 and
16
SCIGS25 to final concentration of 11.9 and 11.3 Cmmol l-1 respectively, representing 63
17
Cmol % of the total sesquiterpene produced by these strains. Product to substrate yield of the
18
different compounds are reported in table I and range from 0.0037 for the most abundant
19
sesquiterpene α-santalene to the lowest 0.0003 (Cmmol Cmmol-1) for farnesene in strain
20
SCIGS24 and from 0.0033 to 0.0003 (Cmmol Cmmol-1) in strain SCIGS25.
21
Effect of mixed (glucose/ethanol) substrate feed on sesquiterpene production
22
Under fully oxidative condition S. cerevisiae is capable of simultaneous consumption of
23
glucose and ethanol as substrate. The ratio between the two carbon sources determines
24
changes in the metabolic flux in the central carbon metabolism. During growth on glucose
25
alone has carbon source sesquiterpene are obtained directly from the cytosolic acetyl-CoA
12
1
produced through pyruvate decarboxylase. Under glucose/ethanol aerobic carbon limited
2
cultures if the ethanol content in the feed is maintained below 0.43 Cmol Cmol-1 the ctytosolic
3
acetyl-CoA is derived directly from the ethanol (de-Jong-Gubbels et al., 1995). Here, we
4
investigate the effect of ethanol as alternative substrate for the production of sesquiterpene.
5
To determine if this approach can be applied for improving the production a mixture of
6
0.57:0.43 glucose:ethanol was used as a feed during the RQ controlled fed-batch process
7
previously described. Oxidation of glucose and ethanol in the selected carbon molar ratio
8
through respiratory metabolism would lead to a respiratory coefficient equal to 0.82.
9
Consistently with the previous experiments the RQ controller was set to maintain an RQ value
10
(RQ 0.82), that prevent the occurrence of the fermentative metabolism. The effects of the
11
mixed feed on the fed-batch process are represented in figure 4. Similar to the process on
12
glucose both strains SCIGS24 and SCIGS25 displayed comparable behavior. The RQ value
13
increase until selected set-point during a first phase (15-16 hours) after that was feedback
14
controlled and remains stable around a value of 0.8. When compared to the situation with
15
glucose alone the feed profile of the volume added on time, differs only in minor amount from
16
the theoretical input. However after 16 hours of feed a second phase, were part of the ethanol
17
content in the feed but not glucose was not completely consume occur (Fig. 4). The ethanol
18
fraction that accumulate increase progressively during the cultivation reaching 54%-58% of
19
the ethanol fed and corresponding to 22%- 24% (in Cmol basis) of the total carbon source at
20
the end of the feed for strain SCIGS24 and SCIGS25 respectively. The effects of the mixed
21
substrate on sesquiterpene production are summarized in table I. Consistently with the
22
previous results on glucose α-santalene was the major product (Fig. 5). Ethanol has a positive
23
effect on the sesquiterpene production on both strains. Strains SCIGS24 and SCIGS25
24
displayed a 14-21% increase α-santalene yield with the best yield of 0.0043 Cmmol Cmmol-1
25
of strain SCIGS24. Interestingly FOH did not change significantly in strain SCIGS24 under
13
1
both conditions whereas a 15% improvement was observed on strain SCIGS25 resulting the
2
highest value observed in all process. The yield of the two minor products trans-α-
3
bregamotene and E,E-farnesene was slightly reduced in both strains under these conditions.
4
When ethanol was used as precursor source for sesquiterpene the total yield was improved for
5
both strains SCIGS24 and SCIGS25 of 4.7 and 13.6 % respectively.
6
If we compare the production performances of the two process, using mixed substrate was
7
possible to increase the total sesquiterpene productivity for strain SCIGS24 of 49% from
8
0.038 to 0.076 Cmmol (g biomass)-1 h-1 (Fig. 6), a minor effect was observed in strain
9
SCIGS25 with 24% increase in productivity from 0.036 to 0.048 Cmmol (g biomass)-1 h-1.
10
The best performance was archived using strain SCIGS24 combined with mixed substrate
11
glucose ethanol yielding to a final sesquiterpene concentration of 21.2 Cmol l-1 after 28.5 h of
12
feed (table I).
13
Intracellular sesquiterpene accumulation
14
Time course analysis of the two process shown that in both cases after c.a. 16 hours of feed
15
time a limitation phase occur leading either a slower consumption during the process with
16
glucose only or incomplete carbon consumption with ethanol accumulation in the process
17
with mixed glucose/ethanol. As direct consequence a decrease in sesquiterpene production in
18
the organic layer was detected (Fig.3 and Fig 5). Several factors could cause the above-
19
described reduction in production performance. In order to investigate if the efflux of
20
sesquiterpene from the cell was related to this effect, intracellular sesquiterpene assay was
21
performed. Only two of the four products detected in the organic layer accumulated
22
intracellularly α-santalene and trans-α-bergamotene. Intra and extracellular product
23
accumulation versus feed time for the two different strains under the two different processes
24
are reported in figure 7. Typically the intracellular fraction of α-santalene and trans-α-
25
bergamotene measured were between 15-18% and 17-29% of the total amount detected
14
1
respectively. It is interesting to notice that the two compound displayed different intracellular
2
accumulation profile. Intracellular trans-α-bergamotene amount increase on time and mainly
3
follow the exrtracellular amount accumulated and was retained in the cell in a proportional
4
amount respect to the extracellular concentration. Whereas the intracellular α-santalene
5
concentration reach saturation around a value of 2 Cmmol l-1, corresponding approximately to
6
the time where the entire process became limited. This phenomenon was detected in both
7
strains and seems to be independent of the carbon source utilized as substrate.
8
9
Discussion
10
Develop of an effective bioprocess: impact of the RQ control on the fed-batch process
11
The developments of an efficient bioprocess for production of sesquiterpene that can compete
12
with the traditional industrial methods require maximization of yield and productivity.
13
Optimization of production performance needs to fast obtain active cell minimizing side
14
product formation. Process design requires detailed knowledge of the product formation in
15
order to achieve optimal balance. In our sesquiterpene engineered strain the extracellular
16
glucose concentration trigger the switch between the flux toward the product formation
17
(Scalcinati et al., 2012a; Scalcinati et al., 2012b) it is therefore critical carefully control the
18
cell metabolism to obtain the best productivity during the entire process. In an initial stage of
19
process optimization we implemented an RQ feedback control method to a programmed
20
limited exponential feed profile of glucose. The RQ control is a physiological type of
21
controller widely applied to improve fed-batch cultivation performances (Kiss et al., 1991;
22
Xiong et al., 2010; Xiao et al., 2006; Bideaux et al., 2006). Here, the feedback control
23
permitted to constantly operate below the critical required value resulting in a robust feed
24
strategy that allow to alleviate glucose repression and Crabtree effect and determine an
25
optimal feed policy. The applied respiratory control systems result in the maximum fed rate
15
1
sustainable from the cell without byproduct formation. The fed batch process was integrated
2
with a consolidated in situ product removal (ISPR) system (Freeman et al., 1993). Once the
3
target compounds are secreted outside the cell, ISPR allow instant and continuous product
4
recovery, sequestering the products in the organic layer. Extracellular sesquiterpene analysis
5
results in the detection of four different products, beside the target compound α-santalene and
6
the know side product FOH originated from the spontaneous dephosporilation of FPP (Wang
7
et al., 2011), minor product trans-α-bergamote and E,E-franesene were detected. Both this
8
compounds are secondary product of the santalene syntase (M. Shalk personal
9
communication). Apparently in the tested conditions santalene synthase has a different level
10
of specificity toward α-santalene confirming the high degree of plasticity often reported for
11
this class of enzymes (Yoshikuni et al., 2006).
12
This methods result in a process capable of producing 12 Cmmol l-1 (163 mg l-1) of α-
13
santalene and over 18 Cmmol l-1 (261 mg l-1) of total sesquiterpene in 30 h of feed. The minor
14
physiological differences obtained within the two engineered strain tested confirm the
15
applicability and reproducibility of the methods.
16
Ethanol as alternative C-source for sesquiterpene production, effect of the mixed
17
substrate
18
In order to obtain a sesquiterpene factory efficient precursors supply is necessary. In yeast
19
sesquiterpene are originated exclusively in the MVA pathway and use cytosolic acetyl-CoA
20
as primary metabolite precursor. Carbon source employed and type of metabolism are known
21
to affect the cellular level of cytosolic acetyl-CoA (Frick et al., 2005; Seker et al., 2005).
22
Previous studies demonstrated that the mevalonate pathway is subject to glucose repression
23
and growth on ethanol lead to an increase in the cytosolyc acetyl-CoA and acetoacetyl-CoA
24
(Quain et al., 1979; Seker et al., 2005). On this bases ethanol represent an attractive carbon
25
source for sesquiterpene production in order to increase the supply of acetyl-CoA for the
16
1
mevalonate pathway. In this prospective, ethanol and ethanol/glucose mix has successfully
2
been applied to increase production of sesquiterpene in yeast (Westfall et al. 2011; Tsuruta et
3
al., 2009). However, although market price of ethanol has decrease substantially due the
4
advent of bio-ethanol a mixed cultivation would rather be economically favorable compared
5
to a fully based ethanol feed. Under respiratory conditions the co-consumption of the two
6
sources is highly regulated by the ratio of the two compounds (de Jong-Gubbels et al., 1995).
7
Herewith, it was possible to use ethanol as direct precursor for sesquiterpene formation and
8
glucose for biomass intermediate biosynthesis. Theoretically on Cmol basis α-santalene
9
maximum achievable yield on ethanol (0.83 Cmol Cmol-1) is 27% higher than glucose (0.56
10
Cmol Cmol-1). Using mixed substrate we further enhance yield and productivity resulting in a
11
final production of 14 Cmmol l-1 (193 mg l-1) of α-santalene and over 21 Cmmol l-1 (292 mg l-
12
1
13
applied to increase precursor supply, result in a rapid approach to improve of the process
14
performances
15
Intracellular accumulation and potential derived toxicity
16
Detailed process characterization show the impossibility to maintain a constant growth rate
17
during the entire process. Independently form the carbon source employed the process was
18
mainly divided into two distinct phase. An initial phase with higher growth rate and a second
19
one with slower growth rate were the process became limited. The previously shown
20
dependency of the productivity on the growth rate was reflected in a reduction of the product
21
formation rate resulting in a suboptimal process. In order to investigate the possible causes of
22
the growth impediment we assay the intracellular content of sesquiterpens. Only two of the
23
four different sesquiterpene produced, α-santalene and trans-α-bergamotene were detected
24
intracellularly with α-santalene representing the main intracellular compound. These two
25
compounds differ substantially from the other linear hydrocarbon E,E-farnesol and E,E-
) of total sesquiterpene in 28.5 h of feed. The multiple carbon source utilization strategy
17
1
farnesene because of their multiple cyclic structure. Previous study has proposed that cyclic
2
terpenes can accumulate into the membrane (Sikkema et al., 1994). It is then reasonable to
3
assume that the structural characteristic might be the reason of the difference observed in the
4
capacity to retain these compounds inside the cell. The export of α-santalene from the cell
5
seems to be the main cause of the inhibitory effect detected that limits its productivity. Up to
6
date the mechanisms of secretion of hydrocarbons are not known. In the natural producers
7
organisms terpenes compounds accumulate into a wide variety of specific multicellular
8
secretory structures (Gershenzon et al., 1994 and Besser et al., 2009) but the mechanism of
9
accumulation has not been fully investigated and very few information are available regarding
10
the secretion process. Due to the hydrophobic nature of these compounds, it has been
11
suggested that the excretion can proceed as simple diffusion without require tranporters
12
(Muramatsu et al., 2008). However the saturation kinetics on time observed for santalene
13
might suggest that yeast utilize unspecific transporter to secrete this compound. In yeast
14
strains engineered for production of different sesquiterpene production, transcriptional up
15
regulation of multidrug transporters related genes resulted induced (Verwaal et al., 2010; Ro
16
et al., 2008). Consistently, our transcriptome studies performed in presence of high
17
sesquiterpene concentration were characterized by the overexpression pleoitropic drug
18
resistance (PDR) network genes (unpublished result).
19
Nowadays, there are only few evidences regarding toxicity of plan terpenes compound on S.
20
cerevisiae (Parveen et al., 2004; Sikkema et al., 1995). Indeed, intra-membrane accumulation
21
of cyclic sesquiterpene has been shown to produce toxic effect caused by loss of membrane
22
integrity (Sikkema et al., 1995). The observed physiology obtained can be explained with the
23
excessive intracellular α-santalene accumulation and consequent toxicity related effect.
24
Further investigation would be necessary in order to fully understand and overcome the limits
18
1
of the process in order to achieve production rates with commercial potential for a viable
2
biobased production.
3
4
Acknowledgments
5
This work has been financed by Firmenich SA, the Knut and Alice Wallenberg Foundation
6
and Chalmers Foundation.
7
19
1
2
3
References
Besser K, Harper A, Welsby N, Schauvinhold I, Slocombe S, Li Y, Dixon RA, Broun P. 2009.
4
Divergent regulation of terpenoid metabolism in the trichomes of wild and cultivated
5
tomato species. Plant Physiol 149:499-514.
6
Bideaux C, Alfenore S, Cameleyre X, Molina-Jouve C, Uribelarrea JL, Guillouet ES. 2006.
7
Minimization of glycerol production during the high-performance fed-batch ethanolic
8
fermentation process in Saccharomyces cerevisiae, using a metabolic model as
9
prediction tool. Appl Environ Microbiol. 72:2134-2140.
10
11
12
13
Bohlmann J, Keeling CI. 2008. Harnessing plant biomass for biofuel and biomaterial.
Terpenoid biomaterials. Plant J. 54:656-669.
Chandran SS, Kealey JT, Reeves CD. 2011. Microbial production of isoprenoids. Process
Biochem. 46:1703-1710.
14
Daviet L, Schalk M. 2010. Biotechnology in plant essential oil production: progress and
15
perspective in metabolic engineering of the terpene pathway. Flavour Fragr J 25:123-
16
127.
17
De-Jong-Gubbels P, vanrolleghem P, Heijnen S, van Dijken JP, Pronk JT. 1995. Regulation
18
of carbon metabolism in chemostat cultures of Saccharomyces cerevisiae growth on
19
mixtures of glucose and ethanol. Yeast 11:407-418.
20
21
22
23
Dellomonaco C, Fava F, Gonzales R. 2010. The path to next generation biofuels: successes
and challenges in the era of synthetic biology. 9:3.
Freeman A, Woodley JM, Lilly MD. 1993. In situ product removal as tool for bioprocessing.
Nat Biotechnol 11:1007-1012.
24
Frick O, Wittmann C. 2005. Cahrachterization of the metabolic shift between oxidative and
25
fermentative growth in Saccharomyces cerevisiae by comparative 13C flux analysis.
26
Microb Cell Fact 4:30.
20
1
2
Gershenzon J. 1994. Metabolic cost of terpenoid accumulation in higher plants. J Chem Ecol
20:1281-1328.
3
GeurtsTG, de Kok HE, Roles JA. 1980. A quantitative description of the growth of
4
Saccharomyces cerevisiae CBS 426 on amixed substrate of glucose and ethanol.
5
Biotechnol Bioeng 22: 2031-2043.
6
7
8
9
10
11
12
13
14
15
Hong K, Nielsen J. 2012. Metabolic engineering of Saccharomucse cerevisiae a key cell
factory platform for future biorefineries. Cell Mol Life Sci. in press
Keasling JD. 2012. Synthetic biology and development tools for metabolic engineering.
Metab Eng. in press.
Keasling JD. 2010. Manufacturing molecules through metabolic engineering. Science
330:1355-1358.
Kim IK, Roldao A, Siewers V, Nielsen J. 2012. A systems-level approach for metabolic
engineering of yeast cell factories. FEMS Yeast Res 12:228-248.
Kiss RD, Stephanopoulos G. 1991. Metabolic activity control of the L-lysine fermentation by
restrained growth fed-batch strategies. Biotechnol Prog 7:501-509.
16
Lee SK, Chou H, Ham TS, Lee TS, Keasling JD. 2008. Metabolic engineering of
17
microorganisms for biofuel production: from bugs to synthetic biology to fuels. Curr
18
Opin Biotechnol 19:556-563.
19
20
Lee J, Lee SY, Park S, Middelberg APJ. 1999. Control of fed-.batch fermentations.
Biotechnol Adv 17:29-48.
21
Leonard E, Ajikumar PK, Thayer K, Xiao W, Mo JD, Tidor B, Stephanopoulos G, Prather
22
KLJ. 2010. Combining metabolic and protein engineering of terpenoid biosynthetic
23
pathway for overproduction and selectivity control. Proc Natl Acad Sci 107:13654-
24
13659.
21
1
Maury J, Asadollahi M A, Møller K, Clark A, Nielsen J. 2005. Microbial isoprenoid
2
production: an example of green chemistry through metabolic engineering. Adv
3
Biochem Eng Biotechnol 100:19-51.
4
5
McCaskill D, Croteau R. 1997. Prospects for bioengineering of isoprenoid biosynthesis. Adv
Biochem Eng/Biotechnol. 55:107-146.
6
Muramatsu M, Ohto C, Obata S, Sakuradani E, Shimizu S. 2008. Various oil and detergent
7
enhance the microbial production of farnesol and related prenyl alcohols. 106:263-268.
8
Nielsen J, Keasling JD. 2011. Synergies between synthetic biology and metabolic
9
10
11
12
13
engineereing. Nat Biotechnol 29:693-695.
Nielsen, J Villadsen, J Liden G. Bioreaction Engineering Principles 2ed. 2003. , New York:
Kluywer Plenum.
Otero JM, Panagiotou G, Olsson L. 2007. Fueling industrial biotechnology growth with
bioethanol. Adv Biochem Eng Biotechnol 108:1-40.
14
Parveen M, Hasan MK, Takahashi J, Murata Y, Kitagawa E, Kodama O, Iwahashi H. 2004.
15
Response of Saccharomyces cerevisiae to monoterpene: evaluation of antufungal
16
potential by DNA microarray analysis. J Antimicrob Chemoth 54:46-55.
17
18
Peralta-Yahya PP, Ouellet M, Chan R, Mukhopadhyay A, Keasling JD, Lee TS. Identification
and microbial production of a terpene-based advanced biofuel. Nat Comm 2:483.
19
Pham HTB, Larsson G, Enfors SO. 1998. Growth and energy metabolism in aerobic fed-batch
20
cultures of Saccharomyces cerevisiae: simulation and model verification. Biotechnol
21
Bioeng 60: 474-482.
22
Quain DE, Haslam JM. 1979. The effects of catabolite derepression on the accumulation of
23
steryl esters and the activity of ß-hydroxymethylglutaryl-CoA reductase in
24
Saccharomyces cerevisiae. J Gen Microbiol 111:343-351.
22
1
2
Renninger NS, Newman JD, Reiling KK. 2008. Fuel components, fuel compositions and
methods of making and using same. US2008/0092829 A1.
3
Ro DK, Ouellet M, Paradise EM, Burd H, Eng D, Paddon CJ, Newman JD, Keasling JD.
4
2008. Induction of multiple pleiotropic drug resistance genes in yeast engineered to
5
produce an increased level of anti-malarial drug precursor, artemisinic acid. BMC
6
Biotechnol 8:83.
7
Ro DK, Paradise EM, Ouellet M, Fisher KJ, Newman KL, Ndungu JM, Ho KA, Eachus RA,
8
Ham TS, Kirby J, Chang MCY, Withers ST, Shiba Y, Sarpong R, Keasling JD. 2006.
9
Production of the anitmalarial drug precursor artemisinic acid in engineered yeast.
10
11
12
Nature 440:940-943.
Rude MA, Schirmer A. 2009. New microbial fuels: a biotech perspective. Curr Opin
Microbiol. 12: 274-281.
13
Scalcinati G, Knuf C, Partow S, Chen Y, Maury J, Schalk M, Daviet L, Nielsen J, Sewers V.
14
2012a. Dynamic control of gene expression in Saccharomyces cerevisiae engineered
15
for the production of plant sesquiterpene α-sanatlene in a fed-batch mode. Metab Eng
16
14:91-103.
17
Scalcinati G, Partow S, Siewers V, Schalk M, Daviet L, Nielsen J. 2012b. Combined
18
metabolic engineering of precursor and co-factor supply to increase α-santalene
19
production by Saccharomyces cerevisiae…………
20
Seker T, Moller K, Nielsen J. 2005. Analysis of acyl CoA ester intermediates of the
21
mevalonate pathway in Saccharomyces cerevisiae. Appl Microbiol Biotechnol 67:119-
22
124.
23
24
Sherman F, Fink GR, Hicks JB.1986. Methods in yeast genetics – Laboratory Manual. New
York: Cold Spring Harbor.
23
1
Siddiqui MS, Thodey K, Trenchard I, Smolke CD. 2012. Advancing secondary metabolite
2
biosynthesis in yeast with synthetic biology tools. FEMS Yeast Res 12:144-170.
3
Sikkema J, de Bont JAM, Poolman B. 1995. Mechanisms of membrane toxicity of
4
5
6
7
8
9
10
11
hydrocarbons. Microbiol Rev 59:201-222.
Sikkema J, de Bont JAM, Poolman B. 1994. Interaction of cyclic hydrocarbons with
biological membranes. J Biol Chem 18:8022-8028
Stephanopoulos G. 2007. Challenges in engineering microbes for biofuels production. Science
315: 801-804.
Tsuruta H, Lenihan JR, Rika R. 2009. Production of isoprenoids. US 2009/01377014 A1.
Van Gulik WM, Heijnen JJ. 1995. A network stoichiometry analysis of microbial growth and
product formation. Biotechnol and Bioeng 48:681-698.
12
Verduyn V, Postma E, Scheffers W.A, Van Dijken JP. 1992. Effect of benzoic acid on
13
metabolic fluxes in yeasts: A continuous-culture study on the regulation of respiration
14
and alcoholic fermentation. Yeast 8:501-517.
15
Verwaal R, Jiang Y, Daran JM, Sandmann G, van den Berg J, van Ooyen AJJ. 2010.
16
Heterologous carotenoid production in Saccharomyces cerevisiae induces the
17
pleiotropic drug resistance stress response. Yeast 27:985-988.
18
19
Wang C, Kim JY, Choi ES, Kim SW. 2011. Microbial production of farnesol (FOH): current
states and beyond. Proc Biochem 46:1221-1229.
20
Westfall P, Pitera DJ, Lenihan JR, Eng D, Woolard FX, Regentin R, Horning T, Tsuruta H,
21
Melis DJ, Owens A, Fickers S, Diola D, Benjamin KR, Keasling JD, Leavell MD,
22
McPhee DJ Renninger NS, Newman JD, Paddon CJ. 2011. Production of
23
amorphadiene in yeast, and its conversion to dihydroartemisinic acid, precursor to the
24
antimalarial agent artemisinin. Proc Natl Acad Sci 109:111-118.
24
1
Woehrer W, Roehr M. 1981. Regulatory aspects of bakers’ yeast metabolism in aerobic fed-
2
batch cultures. Biotechnol Bioeng 23:567-581.
3
Xiao J, Shi Z., Gao P, Feng H, Duan Z, Mao Z. 2006. On-line optimization of glutamate
4
production based on balanced metabolic control by RQ. Bioprocess Biosyst Eng
5
29:109-117.
6
Xiong ZQ, Guo MJ, Guo YX, Zhuang YP, Wang NS, Zhang SL. 2001. RQ feedback control
7
for simultaneous improvement of GSH yield and GSH content in Saccharomyces
8
cerevisiae T65. Enz Microb Technol 46:598-602.
9
Yoshikuni Y, Ferrin TE, Keasling JD. 2006. Designed divergent evolution of enzyme
10
11
function. Nature 440:1078-82.
Zhang F, Rodriguez S, Keasling JD.2011. Metabolic engineering of microbial pathway for
12
advanced biofuel production. Curr Opin Biotechnol 22:775-783.
13
14
15
16
17
18
19
Strain
Table I. Titers and yield of target sesquiterpenes obtained during aerobic fed batch cultivation
of strains SCIGS24 and SCIGS25. α-santalene yield (YsSan); E,E-farnesol yield (YsFar); transα-bergamotene (YsBer) and E,E-Farnesene yield (YFse) (Cmmol (Cmmol substrate)-1), referred
to the feed phase. Total sesquiterpene titer (TotSesq) (Cmmol l-1) measured at the end of the
feed process. Value represents the mean ±S.D. of three independent cultivations.
Feed C-source
Glucose
EtOH
Sesquiterpene Yield
YsSan
Cmol l-1
SCIGS24
7.40 ±0.11
4.15 ±0.04
SCIGS25
2.98 ±0.08
7.40 ±0.11
4.15 ±0.04
2.98 ±0.08
YsFar
YsBer
Titers
YsFse
YsTot
TotSesq
Cmmol (Cmmol substrate)-1
Cmmol l-1
0.0037 ±0.0005
0.0018 ±0.0004
0.0004±0.0001
0.0003 ±0.0001
0.0061
18.7 ±0.23
0.0043 ±0.0002
0.0018±0.0001
0.0003 ±0.0001
0.0001 ±0.0001
0.0064
21.2±0.11
0.0033 ±0.0002
0.0017 ±0.0001
0.0004±0.0001
0.0003±0.0001
0.0057
17.9±0.34
0.0042 ±0.0001
0.0020 ±0.0002
0.0002 ±0.0001
0.0001 ±0.0001
0.0066
19.7±0.27
20
21
Figure Legend
22
Figure 1.
25
1
(A) The configuration of the in situ product removal (ISPR) fed-batch RQ controlled
2
cultivation process. A stirred tank reactor is operated in fed batch cultivation mode as double
3
phase system adding organic solvent on top of the culture and feeding concentrated culture
4
medium. The product is continuously captured in the organic phase due to its high
5
hydrophobicity. Feed delivery is designed with an exponential policy and it is controlled
6
through a feed-back loop. Fermentation exhaust gas analysis allow the on line determination
7
of the respiratory quotient, the instant RQ measure modulated the fed addition by a PI
8
controller in order to maintain the desired set point. (B) Scheme of sesquiterpene (C15)
9
synthesis in yeast. The functional isoprene units isopentenyl diphosphate (3) and its isomers
10
dimethylally diphosphate (4) are synthesized from cytosolic acetyl-CoA (1) through
11
mevalonate (2) intermediate in the mevalonate pathway. The universal sesquiterpene
12
precursor farnesyl diphiosphate (5) is subsequently obtained by the condensation of three
13
units of isopentenyl diphosphate. Sesquiterpenes α-santalene (6); trans-α-bergamotene (7);
14
E,E-farnesene (8) and E,E-farnesol (9) are obtained in one step conversion from farnesyl
15
diphosphate.
16
17
Figure 2.
18
RQ based feed-control development of ISPR aerobic glucose limited fed batch cultivation.
19
Production phase of sesquiterpene producing S. cerevisiae strain SCIGS24 (A) and SCIGS25
20
(B). Profiles reported represent the time course of biomass formation (gDCW), RQ measure
21
(CTR/OTR), and comparison between exponential calculated feed profile and experimental
22
value (ml). Data presented are representative of three independent cultures; the error bars
23
represent the standard deviation for three independent cultivations.
24
25
Figure 3.
26
Production phase of aerobic ISPR glucose-limited fed-batch cultivation of S. cerevisiae
27
strains SCIGS24 (A) and SCIGS25 (B). α-santalene, E,E-farnesol, trans-α-bergamotene and
28
E,E-farnesol (Cmmol l-1) product accumulation as function of feeding time. The error bars
29
represent the standard deviation for three independent cultivations.
30
31
Figure4.
32
RQ based feed-control development of ISPR aerobic glucose/ethanol mixed substrate fed-
33
batch cultivation. Production phase of sesquiterpene producing S. cerevisiae strain SCIGS24
34
(A) and SCIGS25 (B). Profiles reported represent the time course of biomass formation
26
1
(gDCW), RQ measure (CTR/OTR), ethanol accumulated (Cmmol l-1) and comparison
2
between exponential calculated feed profile and experimental value (ml). Data presented are
3
representative of three independent cultures; the error bars represent the standard deviation for
4
three independent cultivations.
5
6
Figure 5.
7
Production phase of aerobic ISPR glucose/ethanol mixed substrate fed-batch cultivation of S.
8
cerevisiae strains SCIGS24 (A) and SCIGS25 (B). α-santalene, E,E-farnesol, trans-α-
9
bergamotene and E,E-farnesol (Cmmol l-1) product accumulation as function of feeding time.
10
The error bars represent the standard deviation for three independent cultivations.
11
12
Figure 6.
13
Sesquiterpene productivity in a two-phase partitioned aerobic fed-batch. α-Santalene and E,E-
14
farnesol, trans-α-bergamotene and E,E-farnesene production rate in Cmmol (g biomass)-1 h-1
15
(the C-molar weight are 13.62 g Cmol-1 for α-santalene, trans-α-bergamotene, E,E-farnesene
16
and 14.82 g Cmol-1 for E,E-farnesol). Strains SCIGS24 and SCIGS25 are cultivated using
17
glucose or a mixture of glucose and ethanol as carbon source. Error bars represent the
18
standard deviation from three independent cultivations.
19
20
Figure 7.
21
Extracellular (grey area) and intracellular (black area) sesquiterpene α-santalene and trans-α-
22
bergamotene during RQ based double phase aerobic glucose or glucose/ethanol limited fed
23
batch cultivation of strain SCIGS24 and SCIGS25. (A) α-santalene, strain SCIGS24 glucose
24
feed; (B) trans-α-bergamotene, strain SCIGS24 glucose feed; (C) α-santalene, strain
25
SCIGS25 glucose feed; (D) trans-α-bergamotene, strain SCIGS25 glucose feed; (E) α-
26
santalene, strain SCIGS24 glucose/ethanol feed; (F) trans-α-bergamotene, strain SCIGS24
27
glucose/ethanol feed; (G) α-santalene, strain SCIGS25 glucose/ethanol feed; (H) trans-α-
28
bergamotene, strain SCIGS25 glucose/ethanol feed. Error bars represents the standard
29
deviation from three independent cultivations.
30
27
1
Fig.1
2
3
28
1
Fig.2
2
3
29
1
Fig.3
2
3
30
1
Fig. 4
2
3
31
1
Fig.5
2
3
32
1
Fig.6
2
3
33
1
Fig.7
2
34
PAPER IV
Evolutionary engineering of Saccharomyces cerevisiae for
efficient aerobic xylose consumption.
Scalcinati G, J.M. Otero JM, Van Vleet J, Jeffries TW, Olsson L, Nielsen J.
FEMS Yeast research
65
RESEARCH ARTICLE
Evolutionary engineering of Saccharomyces cerevisiae for
efficient aerobic xylose consumption
Gionata Scalcinati1,2, José Manuel Otero1,2,5, Jennifer R.H.Van Vleet3, Thomas W. Jeffries3,4,
Lisbeth Olsson1,2 & Jens Nielsen1,2
1
Department of Chemical and Biological Engineering, Chalmers University of Technology, Gothenburg, Sweden; 2Center for Microbial
Biotechnology, Department of Systems Biology, Technical University of Denmark, Kgs. Lyngby, Denmark; 3Department of Bacteriology, University
of Wisconsin-Madison, Madison, WI, USA and 4Institute for Microbial and Biochemical Technology, Forest Products Laboratory, Madison,
WI, USA
Correspondence: Lisbeth Olsson,
Department of Chemical and Biological
Engineering, Chalmers University of
Technology, SE-41296 Gothenburg, Sweden.
Tel.: +46 0 31 772 3805; fax:
+46 0 31 772 3801;
e-mail: [email protected]
Present address: José Manuel Otero,
Vaccine and Biologics Process Development,
Bioprocess Research and Development,
Merck Research Labs, West Point, PA, USA
Received 21 December 2011; revised 1
March 2012; accepted 2 April 2012.
DOI: 10.1111/j.1567-1364.2012.00808.x
YEAST RESEARCH
Editor: Hyun Ah Kang
Keywords
directed evolution; metabolic engineering;
xylose; Saccharomyces cerevisiae;
transcriptomics.
Abstract
Industrial biotechnology aims to develop robust microbial cell factories, such
as Saccharomyces cerevisiae, to produce an array of added value chemicals presently dominated by petrochemical processes. Xylose is the second most abundant monosaccharide after glucose and the most prevalent pentose sugar found
in lignocelluloses. Significant research efforts have focused on the metabolic
engineering of S. cerevisiae for fast and efficient xylose utilization. This study
aims to metabolically engineer S. cerevisiae, such that it can consume xylose as
the exclusive substrate while maximizing carbon flux to biomass production.
Such a platform may then be enhanced with complementary metabolic engineering strategies that couple biomass production with high value-added chemical. Saccharomyces cerevisiae, expressing xylose reductase, xylitol dehydrogenase
and xylulose kinase, from the native xylose-metabolizing yeast Pichia stipitis,
was constructed, followed by a directed evolution strategy to improve xylose
utilization rates. The resulting S. cerevisiae strain was capable of rapid growth
and fast xylose consumption producing only biomass and negligible amount of
byproducts. Transcriptional profiling of this strain was employed to further
elucidate the observed physiology confirms a strongly up-regulated glyoxylate
pathway enabling respiratory metabolism. The resulting strain is a desirable
platform for the industrial production of biomass-related products using xylose
as a sole carbon source.
Introduction
Xylose is the most abundant pentose sugar in lignocellulosic feedstocks, including hardwoods and crop residues,
and is the second most abundant monosaccharide after
glucose (Olsson et al., 2004). The demand for industrial
biotechnology processes that leverage sustainable, environmentally favourable and cost-effective raw materials as
alternatives to petrochemical feedstocks is receiving
unprecedented research focus (Otero et al., 2007; Stephanopoulos, 2010). Saccharomyces cerevisiae is a proven,
robust, industrial production platform used for the
expression of a wide range of therapeutic agents, food
and beverage components, added value chemicals
FEMS Yeast Res && (2012) 1–16
(Ostergaard et al., 2000; Chemler et al., 2006; Kim et al.,
2012) and commodity chemicals (e.g. bioethanol) across
large scales (> 50 000 L) (Olsson et al., 2004; Kumar &
Murthy, 2011). Wild-type S. cerevisiae is unable to efficiently utilize xylose as a primary substrate. The field has
largely focused on metabolic engineering of S. cerevisiae for
maximizing carbon flux from xylose to bioethanol under
anaerobic conditions (Chu & Lee, 2007; Cai et al., 2012);
however, the use of S. cerevisae is extended as microbial
cell factory for a variety of added value chemicals (Chemler et al., 2006). The design of a S. cerevisiae platform for
broader biomass-coupled production from xylose would
be favoured without the loss of carbon to overflow metabolites (ethanol, glycerol, xylitol) particularly in the case of
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
2
growth-associated production processes of non-secreted
products that require simultaneous formation of biomass
and target compound (e.g. poly-3-hydroxybutyrate, ß-carotene and lycopene) (Yamano et al., 1994; Tyo et al.,
2010).
Utilization of xylose in yeast and filamentous fungi
occurs by a two-step pathway. First, xylose is reduced to
xylitol via xylose reductase (XR, primarily NADPH consuming), and then xylitol is oxidized to xylulose via xylitol dehydrogenase (XDH, NADH producing) (Wang
et al., 1980). In bacteria, isomerization of xylose to xylulose occurs in a one-step reaction catalysed by xylose
isomerase (Misha & Singh, 1993; Harhangi et al., 2003).
In yeast, fungi and bacteria, the final conversion of xylulose to xylulose-5P via xylulose kinase (ATP consuming)
is conserved.
Recombinant S. cerevisiae strains expressing the Pichia
stipitis xylose reductase (PsXYL1), and P. stipitis xylitol
dehydrogenase (PsXYL2) has lead to transformants that
can oxidatively and exclusively consume xylose, although
resulting in significant xylitol production (Kötter et al.,
1990; Kötter & Ciriacy, 1993; Tantirungkij et al., 1993;
Walfridsson et al., 1995). While over-expression of the
endogenous XKS1 encoding xylulokinase improved the
xylose utilization rate (Ho et al., 1998; Eliasson et al.,
2000; Tiovari et al., 2001), xylitol formation persisted.
There is a redox imbalance, which results from recombinant co-expression of XR and XDH, and because of the
lack of transhydrogenase activity in S. cerevisiae, and
thereby inability to interconvert NADPH and NADH,
there is a surplus formation of NADH and NADP+.
Numerous metabolic engineering efforts employed to alleviate the redox imbalance are discussed above and to further improve the xylose consumption rate have been
reviewed extensively (Amore et al., 1991; Kuyper et al.,
2003, 2004, 2005; Hahn-Hägerdal et al., 2007; van Maris
et al., 2007; Matsushika et al., 2009; Van Vleet & Jeffries,
2009).
Among the possible bottlenecks investigated in xylose
metabolism, several limiting steps have been identified.
The reduced ability of S. cerevisiae to grow efficiently on
xylose has been attributed to: (1) the inefficient xylose
uptake (Amore et al., 1991; Eliasson et al., 2000; Kuyper
et al., 2003, 2004, 2005), (2) the insufficient level of
expression of xylose transporters to enable significant
sugar assimilation (Kötter & Ciriacy, 1993), (3) the redox
imbalance generated in the first two steps of xylose
metabolism involving the XDH and XR from P. stipitis
(Kötter & Ciriacy, 1993; Roca et al., 2003), (4) the level
of aeration (Skoog & Hahn-Hägerdal, 1990; du Preez,
1994; Walfridsson et al., 1995), (5) insufficient pentose
phosphate pathway activity (Kötter & Ciriacy, 1993; Walfridsson et al., 1995) and (6) the inability of pentose
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
G. Scalcinati et al.
sugar metabolism to activate the lower part of glycolysis
(Boles et al., 1993; Müller et al., 1995).
Because of the previously described specificity of XR
for NADPH and XDH for NAD+ and the resulting redox
imbalance, xylose metabolism is partially regulated by the
availability of oxygen in both native and metabolically
engineered yeasts (Skoog & Hahn-Hägerdal, 1990; du
Preez, 1994; Ho et al., 1998). In the presence of oxygen,
excess NADH produced via NAD-dependent XDH can be
respired and the NADPH demand for the XR reaction
provided by the oxidative part of the pentose phosphate
pathway. The level of oxygenation determines the split in
carbon flux between biomass and ethanol production
under aerobic conditions where xylose is mainly converted into biomass, while ethanol production is favoured
under anaerobic conditions (du Preez, 1994). The incomplete respiration of excess NADH under anaerobic conditions leads S. cerevisiae to produce and accumulate
glycerol followed by xylitol. The xylose consumption rate
and the assimilation to biomass increase with increasing
aeration level, relieving the accumulation of NADH, yet
still resulting in glycerol and xylitol formation (Müller
et al., 1995; Jin et al., 2004).
This study aims to metabolically engineer S. cerevisiae
such that it can consume xylose as the exclusive substrate
while maximizing carbon flux to biomass production.
Such a platform may then be enhanced with complimentary metabolic engineering strategies that couple biomass
production with high value-added chemicals. Saccharomyces cerevisiae CEN.PK 113-3C, expressing PsXYL1 (encoding xylose reductase, XR), PsXYL2 (encoding xylitol
dehydrogenase, XDH) and PsXYL3 (encoding xylulose
kinase, XK) from the native xylose-metabolizing yeast
P. stipitis, was constructed, followed by a directed evolution strategy to improve xylose utilization rates. The
resulting strains were physiologically characterized under
aerobic controlled batch fermentations supplemented with
glucose and xylose. Transcriptional profiling was
employed to further elucidate the strain physiology.
Materials and methods
Saccharomyces cerevisiae strain descriptions
All of the strains constructed in this study were derived
from the reference S. cerevisiae strain, CEN.PK 113-7D
(van Dijken et al., 2000). The strains and plasmids used
in this study are listed in Table 1. The strain that was
modified using directed evolution is referred to as
evolved.
Strain CMB.GS001 was derived from the S. cerevisiae
CEN.PK 113-3C wild-type strain. This strain was transformed with the centromeric plasmid pRS314-X123,
FEMS Yeast Res && (2012) 1–16
3
Metabolic enginnering of S. cerevisiae for xylose consumption
Table 1. Saccharomyces cerevisiae strain and plasmid used in this study
Strain or plasmid
CEN.PK 113-7D
CEN.PK 113-3C
CMB.GS001
CMB.GS010†
pRS314-X123
Relevant genotype
Origin/reference
C
MATa URA3 HIS3 LEU2 TRP1 SUC2 MAL2-8
MATa URA3 HIS3 LEU2 trp1-289 SUC2 MAL2-8C
MATa URA3 HIS3 LEU2 TRP1 SUC2 MAL2-8C
pTDH3-PsXYL1 pTDH3-PsXYL2 pTDH3-PsXYL3
MATa URA3 HIS3 LEU2 TRP1 SUC2 MAL2-8C
pTDH3-PsXYL1 pTDH3-PsXYL2 pTDH3-PsXYL3
Evolved
pTDH3-PsXYL1 pTDH3-PsXYL2 pTDH3-PsXYL3 (TRP1, Centromeric)
SRD GmbH*
SRD GmbH*
This study
This study
Haiying et al. (2007)
*Scientific Research and Development GmbH, Oberursel, Germany.
†Single colony isolated after repetitive batch evolutionary process. The three final digits of the strain identifier indicate from which cycle in the
directed evolution the strain originated, with the starting strain referred to as CMB.GS001.
expressing TRP1 encoding for N-(5′ phosphoribosyl)anthranilate isomerase. Into the plasmid pRS314-X123
PsXYL1 encoding xylose reductase (PsXRp), PsXYL2
encoding xylitol dehydrogenase (PsXDHp) and PsXYL3
encoding xylulokinase (PsXKp) all derived from P. stipitis
were cloned under the glyceraldehyde-3-phosphate dehydrogenase (TDH3) constitutive promoter and terminator
(Haiying et al., 2007).
Strains CMB.GS010 were evolved from CMB.GS001
after cycles of repetitive culture selection in shake flasks.
The three final digits of the strain identifier indicate from
which cycle in the repetitive culture the strain originated,
with the starting strain referred to as CMB.GS001 (see
Directed evolution and selection of strain CMB.GS010,
for details).
cultures were prepared following every cycle of repetitive
culture. Culture samples were streaked on plates with the
same selective condition used throughout the evolution
process (minimal media supplemented with 20 g L 1
xylose) and growth at 30 °C. Three randomly selected
single clones were re-streaked once and thereafter grown
in shake flask (see Shake flask cultivation) when the late
exponential phase was reached as determined by biomass
optical density measurements at 600 nm (OD600 nm),
25% (v/v) sterile glycerol was added, and 1.5 mL sterile
cryovials were prepared and stored at 80 °C. From this
final evolutionary cycle of the three isolates, the fastest
growing strain was designed CMB.GS010 and used for
further characterization.
Medium preparation
Yeast strain transformation
Saccharomyces cerevisiae strain CEN.PK113-3C was transformed with plasmid pRS314-X123 (Haiying et al., 2007)
using a traditional lithium acetate treatment (Gietz &
Woods, 2002). Transformants were selected using synthetic dextrose agar plates without tryptophan (ScD-trp).
Directed evolution and selection of strain
CMB.GS010
Mutants of CMB.GS001 with higher specific growth rates
on xylose were selected for by serial transfer of cells using
repetitive cultures in shake flasks. Specifically, a 500-mL
shake flask containing 100 mL of synthetic minimal medium with 20 g L 1 xylose was inoculated with CMB.
GS001. After 60 h, a new shake flask culture having the
same medium composition was inoculated with cells
from the preceding shake flask at an initial OD600 nm of
0.025. This procedure was repeated for four iterations.
Thereafter, the culture time was reduced to 48 h. This
48-h cultivation was repeated for six iterations after
which strain CMB.GS010 was isolated. Cryovials of stock
FEMS Yeast Res && (2012) 1–16
A previously described synthetic minimal medium containing trace elements and vitamins was used for all shake
flasks and stirred tank cultivations (Verduyn et al., 1992).
Tryptophan was supplemented for the cultivations of
CEN.PK113-3C to satisfy the auxotrophy.
The medium used for stirred tank batch cultivations
had the following composition: 5 g L 1 (NH4)2SO4,
3 g L 1 KH2PO4, 0.5 g L 1 MgSO4·7H2O, 1 mL L 1
trace element solution, 1 mL L 1 vitamin solution,
0.5 mL L 1 antifoam 204 (Sigma A-8311) and
1.25 mL L 1 Ergosterol/Tween 80 solution (final concentration 0.01 g L 1 Ergosterol and 0.42 g L 1 Tween 80).
The fermentation medium was pH adjusted to 5.0 with
2 M NaOH and autoclaved. For the cultivations on glucose, the concentration was 20 g L 1, and for the cultivations on xylose, the concentration was 20 g L 1. Both the
sugar solutions were added by sterile filtration using a
cellulose acetate filter (0.20 lm pore size Minisart®-Plus
Satorius AG).
The medium used for shake flask cultivations had the
same composition as described above, but the (NH4)2SO4
concentration was increased to 7.5 g L 1 and the
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
4
KH2PO4 to 14.4 g L 1 together with 20 g L 1 of glucose
or xylose, and the pH was adjusted to 6.5 prior to autoclaving.
A solid synthetic minimal medium containing 5 g L 1
(NH4)2SO4, 3 g L 1 KH2PO4, 0.5 g L 1 MgSO4·7H2O,
1 mL L 1 trace element solution, 1 mL L 1 vitamin solution, supplemented with 20 g L 1 xylose 20 g L 1 agarose
was used to maintain and isolate the evolved mutant.
A yeast extract peptone dextrose (YPD) complex medium was used for yeast growth prior to transformation
with the following composition (g L 1): 10 yeast extract,
20 peptone, 20 glucose and 20 agar.
A synthetic dextrose minus tryptophan medium (ScDtrp) was used as selective media post-transformation with
the following composition (g L 1): 7.25 Dropout powder
(J.T. Baker), 20 agar and 20 glucose.
G. Scalcinati et al.
Analysis
Cell mass determination
The optical density was determined at 600 nm using
spectrophotometer (Shimadzu UV mini 1240). Dry cell
weight measurements were determined as previously
described (Nielsen & Olsson, 1997).
Extracellular metabolite analysis
Extracellular metabolite concentrations were determined
by HPLC as previously described (Eliasson et al., 2000).
Transcriptomics
RNA sampling and isolation
Shake flask cultivation
Cultivations were carried out in 500-mL baffled Erlenmeyer flasks with two diametrically opposite baffles and
side necks for aseptic sampling by syringe. The flasks
were prepared with 100 mL of medium as previously
described and cultivated in a rotary shaker at 150 r.p.m.
with the temperature controlled at 30 °C. The pH of the
medium was adjusted to 6.5 with 2 M NaOH prior to
sterilization.
Stirred tank batch fermentations
Stirred tank cultivations were performed in 2.2 L Braun
Biotech Biostat B fermentation systems with a working
volume of 2 L. The cultivations were operated at aerobic
and anaerobic conditions with glucose or xylose as the
carbon source. The fermenters were integrated with the
Braun Biotech Multi-Fermenter Control System (MFCS)
for data acquisition. The temperature was controlled at
30 °C, and agitation was maintained at 600 r.p.m. Dissolved oxygen was monitored using an autoclavable
polarographic oxygen electrode. During aerobic fermentations, the sparging flow rate of air was 2 vvm (volume
per volume minute). During anaerobic cultivations, nitrogen containing < 5 ppm O2 was used for sparging at a
flow rate of 2 vvm, with < 1% air-saturated oxygen in
the fermenter as confirmed by the dissolved oxygen measurement and the off-gas analyser. The pH was controlled
constant at 5.0 by automatic addition of 2 M KOH. Offgas passed through a condenser cooled to 4 °C to minimize evaporation, and oxygen and carbon dioxide concentrations were determined by the off-gas analyser as
previously described (Christensen et al., 1995). Fermentations were inoculated from shake flask precultures to a
starting OD600 nm 0.01.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
Samples for RNA isolation from the late exponential
phase of glucose-limited and xylose-limited batch, and
continuous cultivations were taken by rapidly sampling
25 mL of culture into a sterile tube with crushed ice.
Cells were immediately centrifuged (5000 g at 0 °C for
2.5 min), the supernatant was discarded and the pellet
was frozen in liquid nitrogen. Total RNA was extracted
using the RNeasy® Mini Kit (Qiagen, Valencia, CA)
according to manufacturer’s instructions. RNA sample
integrity and quality were determined prior to hybridization with an Agilent 2100 Bioanalyzer and RNA 6000
Nano LabChip kit according to the manufacturer’s
instruction (Agilent, Santa Clara, CA).
Probe preparation and hybridization to DNA
microarrays
Messenger RNA (mRNA) extraction, cDNA synthesis,
labelling and array hybridization to Affymetrix Yeast Genome Y2.0 arrays were performed according to the manufacturer’s recommendations (Affymetrix GeneChip®
Expression Analysis Technical Manual, 2005–2006 Rev.
2.0). Washing and staining of arrays were performed
using the GeneChip Fluidics Station 450 and scanning
with the Affymetrix GeneArray Scanner (Affymetrix,
Santa Clara, CA).
Microarray gene transcription analysis
Affymetrix Microarray Suite v5.0 was used to generate
CEL files of the scanned DNA microarrays. These CEL
files were then processed using the statistical language
and environment R v2.9.1 (R Development Core Team,
2007, www.r-project.org), supplemented with BIOCONDUCTOR v2.3 (Biconductor Development Core Team, 2008,
FEMS Yeast Res && (2012) 1–16
5
Metabolic enginnering of S. cerevisiae for xylose consumption
www.bioconductor.org) packages Biobase, affy, gcrma and
limma (Smyth, 2005). The probe intensities were normalized for background using the robust multiarray average
(RMA) method only using perfect match (PM) probes
after the raw image file of the DNA microarray was visually inspected for acceptable quality. Normalization was
performed using the qspline method, and gene expression
values were calculated from PM probes with the median
polish summary. Statistical analysis was applied to determine differentially expressed genes using the limma statistical package. Moderated t-tests between the sets of
experiments were used for pair-wise comparisons. Empirical Bayesian statistics were used to moderate the standard errors within each gene, and Benjamini–Hochberg
method was used to adjust for multi-testing. A cut-off
value of adjusted P < 0.01 (referred to as Padjusted) was
used for statistical significance, unless otherwise specified
(Smyth, 2005). Gene ontology process annotation was
performed by submitting differentially expressed gene
(adjusted P < 0.01) lists to the Saccharomyces Genome
Database GO Term Finder resource and maintaining a
cut-off value of P < 0.01 for hypergeometric testing of
cluster frequency compared to background frequency
(Ball et al., 2000). Successively, the reporter feature algorithm (Patil & Nielsen, 2005) has been applied on the
dataset to identify transcription factor analysis (TFs)
around which the most significant changes occur. Metabolic pathway mapping was performed using Pathway
Expression Viewer of the Saccharomyces Genome Database, where lists of differentially expressed genes
(Padjusted < 0.01, |log-fold change| > 1) between two conditions were submitted (Ball et al., 2001).
Results
Physiological characterization of CMB.GS001
Batch cultivations of the xylose-fermenting S. cerevisiae
strain CMB.GS001 was investigated in synthetic medium
supplemented with 20 g L 1 xylose. In contrast to the
reference strain CEN.PK 113-7D, which cannot grow on
xylose, the recombinant strain grew aerobically on xylose
with a specific growth rate of 0.02 h 1 and a xylose consumption rate of 0.08 g (g dry cell weight) 1 h 1,
< 2 g L 1 xylose was consumed (Fig. 1 and Table 2).
Fig. 1. Time course of aerobic batch culture on defined minimal
medium supplemented with 20 g L 1 xylose of strain CEN.PK113-7D
(empty symbols) and CMB.GS001 (filled symbols). Xylose, (circle)
(g L 1) and biomass (square) (g DCW L 1) concentrations are
presented as the functions of cultivation time. Data represent the
average of three independent cultures.
20 g L 1 xylose. This approach targeted strain selection
based on biomass formation rate, directly coupled to the
xylose consumption rate. After four batch cultures, strain
CMB.GS001 demonstrated an appreciable improvement
in xylose consumption (Fig. 2a). After serial cultivations,
over 10 cycles covering a period of 500 h (21 days) the
xylose consumption for strain CMB.GS010 increased 15fold to 20 g L 1 and the biomass production increased
52-fold to 9.37 (g dry cell weigh) l 1 (Fig. 2a). A total of
74 cell generations were produced across the ten cycles of
directed evolution resulting in a doubling time decrease
of sixfold from 34.7 to 5.42 h, with the final 50–74 generations not yielding any decrease in the doubling time
(Fig. 2b).
In order to investigate the possible causes of the dramatic increase in the xylose consumption of CMB.GS010,
the plasmid was removed and sequenced, but no mutation was detected compared to the original plasmid
pRS314-X123 suggesting that the improved xylose consumption rate is a consequence of mutations in the genome level and not in the plasmid carrying the properties
needed for xylose metabolism.
Physiological characterization of CMB.GS010
Directed evolution of CMB.GS001
Batch xylose fermentation
Directed evolution was applied to select a spontaneous
mutant with higher specific growth rate on xylose. The
constructed xylose-fermenting strain CMB.GS001 was
subjected to repetitive serial transfers in batch shake flask
cultivations with minimal medium supplemented with
Strain CMB.GS010 was physiologically characterized in
aerobic batch fermentations supplemented with 20 g L 1
xylose or 20 g L 1 glucose. Xylose was completely consumed within 60 h with biomass (62% Cmol C-mol 1
xylose) and carbon dioxide (37% Cmol Cmol 1 xylose)
FEMS Yeast Res && (2012) 1–16
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
6
G. Scalcinati et al.
Table 2. Physiological parameters obtained during aerobic batch cultivation of strains CMB.GS001, CMB.GS010 and reference strain CEN.
PK113-7D. Values represent the mean ± SD of two independent fermentation performed in triplicate (n = 3)
Strain
Carbon source
CMB.GS001
Xylose
Specific growth rate (h 1)
0.02 ± 0.02
Sugar consumed (Cmol L 1)
Glucose
–
Xylose
0.07 ± 0.02
Sugar consumption rate [g (g dry cell weight) 1 h 1]
Glucose
–
Xylose
0.08 ± 0.004
Biomass yield (Cmol Cmol 1)
0.48 ± 0.03
Carbon recovery (%)
94.9 ± 3.5
Productivities [g (g dry cell weight) 1 h 1]
n.a.
CO2
Ethanol
n.a.
Xylitol
n.a.
Glycerol
n.a.
Acetate
n.a.
CMB.GS010
Xylose
Glucose
Glucose/Xylose*
CEN.PK113-7D
Glucose†
0.18 ± 0.01
0.34 ± 0.01
0.32 ± 0.01
0.36
–
0.53 ± 0.04
0.64 ± 0.01
–
0.31 ± 0.02
0.05 ± 0.01
0.66
–
–
0.31 ± 0.02
0.62 ± 0.01
100.1 ± 1.1
2.31 ± 0.06
–
0.16 ± 0.02
105.7 ± 2.5
2.62 ± 0.05
0.21 ± 0.03
0.19 ± 0.01
103.6 ± 2.1
2.36
–
0.15
103.7
0.21 ± 0.01
0.02 ± 0.002
0
0.002 ± 0.001
0
0.53
0.76
0
0.10
0.08
0.53
0.79
0
0.17
0.04
0.49
0.96
0
0.18
0.02
± 0.001
± 0.002
± 0.002
± 0.003
± 0.004
± 0.001
± 0.001
± 0.007
n.a., not available, because of the relative minimal growth.
*Values relative to the first phase of growth (phase I Fig. 4) when glucose with a small fraction of xylose are used.
†
Values from Otero JM, unpublished.
Fig. 2. (a) Comparison in xylose consumption (bars) and biomass production (line) during repetitive growth of Saccharomyces cerevisiae CMB.
GS001 in shake flask cultures on synthetic medium with 20 g L 1 xylose. Shake flask generation represents the number of specific shake flasks
in the series of repetitive cultivations performed to select for mutants with higher specific growth rates and xylose utilization rates (for details,
see Directed evolution and selection of strain CMB.GS010, Materials and methods). (b) Doubling time during the serial transfers of S. cerevisiae
in shake flask cultures on synthetic medium with 20 g L 1 xylose as a function of the number of cell generations. Each data point represents the
doubling time of a single shake flask culture estimated from OD600 nm measurements. The small plot in the top right represents all 10 cycles,
noting the initial doubling time of CMB.GS001 of 35 h, and the rapid decrease to 10 h within < 20 cell generations. For cell generations 50–74,
there was no significant improvement in the specific growth rate.
as the major fermentation products, noting the complete
absence of xylitol during the culture (Table 2). The xylose
consumption rate was highest 0.31 g xylose (g dry cell
weight) 1 h 1 when the extracellular xylose concentration
was above 10 g L 1, as demonstrated by the biomass
concentration and peak carbon evolution rate (Fig. 3a)
subsequently decreasing to 0.08 g xylose (g dry cell
weight) 1 h 1 until xylose exhaustion.
To further investigate whether xylose consumption is
sensitive to changes in extracellular xylose concentration,
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
CMB.GS010 was cultivated with synthetic media supplemented with 10 g L 1 xylose. Under this condition, the
strain exhibits a maximum specific growth rate of
0.11 h 1 compared with 0.18 h 1 when supplemented
with 20 g L 1 xylose. The reduced extracellular concentration of xylose to 10 g L 1 resulted in an increased lag
phase (12–24 h) and maximum specific xylose consumption rate of 0.26 g xylose (g dry cell weight) 1 h 1.
Ability of CMB.GS010 to grow under anaerobic condition was tested in batch fermentation conditions with
FEMS Yeast Res && (2012) 1–16
7
Metabolic enginnering of S. cerevisiae for xylose consumption
Fig. 3. Time course of aerobic batch culture on defined minimal medium supplemented with 20 g L 1 xylose (a) and 20 g L 1glucose (b) of
strain CMB.GS010 (evolved strain). For all plots presented, carbon evolution rate (CER) (dashed line), oxygen uptake rate (OUR) (solid line)
(mM h 1), and xylose (circle), glucose (triangle), ethanol (diamond) (g L 1); and biomass (square) (g DCW L 1) concentrations as the functions of
cultivation time. Data represent the average of three independent cultures.
20 g L 1 xylose as the sole carbon source. After 100 h,
no growth or xylose consumption was observed. To
ensure that the absence of growth was a direct consequence of the anaerobic environment, a recovery experiment was performed, where the culture was aerated
quickly from anaerobic to aerobic condition. Growth was
immediately restored to the above-described aerobic
physiology (data not shown).
Batch glucose fermentation
The aerobic and anaerobic physiology of CMB.GS010 was
evaluated in glucose-supplemented batch fermentations to
quantify the possible effects of directed evolution on the
maximum specific growth rate and the product yields
compared to the reference strain CEN.PK113-7D. During
aerobic conditions, strain physiology was comparable to
previous results with CEN.PK 113-7D (Fig. 3b). The
main differences were a small reduction in the maximum
specific growth rate and a fourfold higher acetate production, and a small reduction in ethanol production
(Table 2). Anaerobic cultivation also showed similar
results. Similarly, a reduction in the maximum specific
growth rate and in the ethanol yield was observed (data
not shown).
Batch mixed substrate fermentation
In order to investigate the proprieties of the strain CMB.
GS010 with respect to mixed sugar utilization, the strain
was grown aerobically in a mixture containing 10 g L 1
glucose and 10 g L 1 xylose. The results show that both
sugars were completely consumed, however with glucose
remaining the preferred substrate. Three different growth
phases can be identified (Fig. 4). During the first growth
FEMS Yeast Res && (2012) 1–16
phase (0–21 h), cells consumed 10 g L 1 glucose and
1.6 g L 1 xylose in the same period (16% more carbon
resulting from xylose consumption). The maximum specific growth rate was slightly lower compared with the
growth on glucose only (Table 2). Following glucose
exhaustion, there was a second growth phase (21–32 h)
where the remaining xylose, 8 g L 1 (0.27 Cmol L 1),
was consumed in conjunction with the re-assimilation
of ethanol produced during the glucose consumption
phase. During this phase, 0.17 Cmol L 1 xylose and
0.15 Cmol L 1 ethanol were consumed. In this phase, the
maximum specific growth rate decreased 2.5-fold from
0.32 to 0.13 h 1. The maximum xylose consumption rate
during the first growth phase on glucose was 0.21 g (g
dry cell weight) 1 h 1. Once glucose was depleted, the
maximum xylose consumption rate was 0.18 g (g dry cell
weight) 1 h 1. After ethanol re-assimilation, the xylose
consumption continued until all the sugar was consumed
in the third and final growth phase (> 32 h) with a
reduced maximum consumption rate of 0.06 g (g dry cell
weight) 1 h 1. In contrast to the glucose consumption
phase, the xylose–ethanol phase was characterized by a
large production of biomass, corresponding to a 28%increase in biomass yield (Cmol Cmol 1).
The fermentation characteristics of strain CMB.GS010
were also investigated under anaerobic growth on a medium containing 10 g L 1 glucose and 10 g L 1 xylose.
However, only glucose was fully consumed (data not
shown).
Transcriptome characterization
Transcriptome characterization was performed with the
evolved strain (CMB.GS010) cultivated in batches with
xylose and glucose as carbon sources, and the un-evolved
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
8
Fig. 4. Time course of aerobic mixed substrate batch cultivation on
defined minimal medium supplemented with 10 g L 1 xylose and
10 g L 1 glucose of strain CMB.GS010 (evolved strain). Carbon
evolution rate (CER) (dashed line), oxygen uptake rate (OUR) (solid
line) (mM h 1), and xylose (circle), glucose (triangle), ethanol
(diamond) (g L 1) and biomass (square) (g DCW L 1) concentrations
as functions of cultivation time. Vertical line identifies the three
different growth phase (I, II and II). Data represent the average of
three independent cultures.
strain (CMB.GS001) with glucose as the sole carbon
source in batch cultivations. This different cultivation
conditions were selected to elucidate overall carbon flux
distributions observed in CMB.GS010 compared to CMB.
GS001. The specific comparisons made were focused on
identifying fermentative vs. respiro-fermentative metabolism for growth on the different carbon sources. Similar
to other studies, the direct comparison with the unevolved strain on xylose was not possible because of the
poor and un-exponential growth on minimal media
(Salusjärvi et al., 2006).
Principal component analysis (PCA) of the expression
data after normalization showed that the evolved strain
grown on xylose in batch conditions clustered with clear
separation from the evolved and un-evolved strain grown
on glucose (data not shown). Gene ontology (GO) term
enrichment analysis from the significant differential gene
expressions between the evolved strain (CMB.GS010)
grown on xylose or glucose and the un-evolved (CMB.
GS001) grown on glucose was performed. A schematic
representation of the GO process term identified (Pvalue < 0.01) is represented in Fig. 5a. The terms ‘citrate
cycle (TCA cycle)’, ‘glyoxylate and dicarboxylate metabolism’, ‘peroxisome’ and ‘oxidative phosphorylation’ were
among the most represented functional category with
respect to metabolism of the evolved strain on xylose compared with the un-evolved or evolved strains grown on
glucose. In contrast, processes related to biosynthesis of
several amino acids were repressed. TF analysis was used
to analyse whether the evolved mutant physiology was
affected at a global regulatory level. Figure 5b represents
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
G. Scalcinati et al.
the increase in the expression of genes regulated by different TFs. The main carbon catabolite repressor regulator
SNF1 and several of its known targets PIP2, OAF1, CAT8
and MIG1, the carbon source responsive ADR1 and the
four subunits of the global respiratory regulator HAP
complex (HAP1, HAP2, HAP3 and HAP4) were among
the identified over-represented TFs in the evolved strain
grown on xylose vs. the un-evolved strain grown on glucose. Interestingly, expression of genes regulated by TFs
linked to the glucose sensor RGT2, such as STD1, and
SNF1 target TFs like INO2 and INO4 (involved in the regulation of lipid metabolism) were down-regulated when
the evolved and un-evolved strains were compared on glucose. The GO and TF results were consistent with the differences in metabolism expressed by the evolved strain
grown on the two sugars. Further, detailed analysis was
performed to analyse the change of gene expression at the
metabolic pathway level (Fig. 6). The significant mRNA
up-regulation of genes encoding enzymes of the central
carbon metabolism often correlating with respiration in
the TCA cycle and glyoxylate pathways correlates well with
the physiological observations that growth on xylose is
dominated by respiratory metabolism. The glyoxylate
pathway (ICL1, MLS1, MDH2, AGX1 encoding isocitrate
lyase, malate synthase, malate dehydrogenase and glyoxylate aminotransferase) was significantly up-regulated in
the evolved strain grown on xylose compared to the
evolved strain grown on glucose or the un-evolved strain
grown on glucose. This pathway had a significantly higher
log-fold change than succinate dehydrogenase, a-ketoglutarate dehydrogenase and succinyl-CoA ligase (SDH1,
SDH2, SDH3, SDH4, KGD1, KGD2 and LSC2, respectively), suggesting that it plays an important role during
xylose respiratory metabolism of S. cerevisiae as found
from studies in chemostat cultures (Regenberg et al.,
2006). Finally, IDP2 and IDP3 (encoding two isocitrate
dehydrogenase) were up-regulated significantly in batch
xylose cultivations with the evolved strain (Fig. 6).
The evolved strain, when cultivated on xylose in a
batch mode, is able to utilize the glyoxylate bypass to efficiently respire the carbon source. Similar to previous
report, the up-regulation of HXK1 and GLK1 (encoding
hexokinase isoenzyme 1 and glucokinase) supports the
hypothesis that xylose is identified as a non-fermentable
carbon source and therefore respired (Herrero et al.,
1995; Jin et al., 2004; Salusjärvi et al., 2008; Runquist
et al., 2009). Furthermore, the expression levels of
MDH2, PCK1 and FBP1 (encoding malate dehydrogenase, phosphoenolpyruvate carboxykinase and fructose1,6-bisphosphatase, respectively) were up-regulated in
the evolved strain cultivated on xylose compared to the
evolved or un-evolved strain cultivated on glucose,
indicating some glyconeogenic activity. It should be
FEMS Yeast Res && (2012) 1–16
Metabolic enginnering of S. cerevisiae for xylose consumption
9
Fig. 5. Integrated analysis of gene ontology (GO) process terms (a) and TFs analysis (b), the cluster frequency is presented on the y-axis.
Compared conditions are the evolved strain CMB.GS010 strain cultivated on batch xylose (XEM_BX), evolved strain CMB.GS010 cultivated in
batch glucose (XEM_BG) and the un-evolved strain CMB.GS001 cultivated on batch glucose (WT_BG). Colour key indicates the different
expression in log-fold change (Padjusted < 0.01).
mentioned though that these genes are also up-regulated at low dilution rates in glucose-limited chemostat
cultures (Regenberg et al., 2006), and expression of
these genes may therefore be associated with respiratory
metabolism.
FEMS Yeast Res && (2012) 1–16
The mRNA expression profile of the evolved strain cultivated on xylose suggests a strong flux towards glucose6-phosphate, requiring inspection of the pentose phosphate pathway (PP pathway). The comparison of differential gene expression in the PP pathway is represented
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
10
G. Scalcinati et al.
Fig. 6. Gene expression levels of central carbon metabolic pathways: tricarboxylic acid (TCA) cycle, glyoxylate pathway, glutamine/glutamate
synthesis and pentose phosphate (PP) pathway are presented. The comparative conditions evaluated include CMB.GS010 cultivated on batch
xylose vs. CMB.GS010 cultivated on batch glucose (Evolv X/Evolv G), shown on the left side box; CMB.GS010 cultivated on batch xylose vs.
CMB.GS001 cultivated on batch glucose (Evolv X/Unevolv G), shown on the right side box. The log-fold change of significantly differentially
expressed genes (Padjusted < 0.01, |log-fold change| > 1) is indicated inside each box next to the gene name; boxes are coloured according to logfold colour scale. If no gene is shown for a given comparative condition, then no significant differential expression changes were detected. The
terms evolved and CMB.GS010, and un-evolved and CMB.GS001, are used interchangeably. The pathway intermediate abbreviations are the
follows: G-6-P, glucose-6-phosphate; F-6-P, fructose-6-phosphate; F-1,6-P, fructose-1,6-biphosphate; GA-3P, glyceraldehyde-3-phosphate; DHAP,
dihydroxyacetonephosphate; 3-PG, 3-phospho-glycerate; PEP, phosphoenol-pyruvate; PYR, pyruvate; AcCoA, acetyl-CoA; Isoct, isocitrate; AKG,
alpha-keto-glutarate; SucCoA, succinyl-CoA; Fum, fumarate; Mal, malate; Oaa, oxaloacetate; Ru-5-P, ribulose-5-phosphate; X-5P, xylulose-5phosphate; S-7-P, sedoheptulose-7-phosphate; E-4-P, erythrose-4-phosphate.
in Fig. 6. Independent of whether the evolved strain was
cultivated on batch xylose or glucose, the SOL genes were
significantly up-regulated (SOL3 and SOL4 encode 6phosphogluconolactonase) compared to growth on glucose. The evolved strain cultivated on xylose, when compared to the evolved and un-evolved on glucose,
exhibited significant up-regulation of transketolase
(encoded byTKL2, log-fold change 4.59 and 3.96). Transketolase (encoded by major isoform TKL1 and minor isoform TKL2) in combination with transaldolase (encoded
by TAL1) enables a reversible link between the non-oxidative PP pathway and glycolysis, allowing the cells to
adapt their NADPH production and ribose-5-phosphate
production to biomass demands (Senac & Hahn-Hägerdal, 1991). The over-expression of TAL1 and TKL1 in
S. cerevisiae over-expressing PsXYL1 and PsXYL2 has been
previously demonstrated, and there was no influence on
growth under either aerobic or anaerobic fermentation
conditions in the TKL1 over-expressed mutant (Walfridsª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
son et al., 1995). TKL2 over-expression was not considered, and the authors concluded that transaldolase
expression in S. cerevisiae is insufficient for the effective
utilization of PP pathway metabolites (Walfridsson et al.,
1995). Furthermore, TKL2 up-regulation has been correlated with carbon-limited chemostat culture (Boer et al.,
2003), but this was not observed here.
Discussion
The strain constructed in this work (CMB.GS010) was
obtained through a combination of genetic modification
(plasmid introduction) and the application of selective
pressure (shake flask repetitive cultivation). Because of
the native inability of S. cerevisiae to metabolize xylose,
three essential genes for xylose uptake from P. stipitis
were introduced. Using repetitive batch cultivation technique, a strain capable of fast aerobic xylose metabolism
was obtained in a relatively short period of time. The
FEMS Yeast Res && (2012) 1–16
Metabolic enginnering of S. cerevisiae for xylose consumption
10-fold increase in the specific growth rate on xylose
under aerobic growth conditions in only 21 days through
repetitive shake flask cultures is the evidence of the efficiency and simplicity of the method. Different evolutionary studies conducted using nutrient limitation as
selective pressure has highlighted the high level of adaptability of S. cerevisiae (Ferea et al., 1999; Sonderegger &
Sauer, 2003; Jansen et al., 2005; Kuyper et al., 2005; Pitkänen et al., 2005). The evolutionary profile observed in
this study during the selection period displayed a rapid
adaptation to the new fast xylose growing condition
rather than a gradual process. Similarly, recent studies
reported that when strong selective pressure is applied to
yeast cultures in laboratory conditions, adaptation occurs
in few steps involving only limited number of mutation
(Hong et al., 2011) and the early phase of evolution plays
a critical role in the adaptation process (Gresham et al.,
2008). Furthermore, plasmid recovery confirms that the
genetic modifications during adaptive evolution are present chromosomally in the host rather than any modifications in the plasmid. The S. cerevisiae strain CMB.GS010
exhibited a specific growth rate and a xylose consumption
rate among the highest reported for S. cerevisiae strains
metabolically engineered for xylose assimilation with XR,
XDH and XK genes and a xylose consumption rate on
minimal media under aerobic conditions (Sonderegger &
Sauer, 2003; Wahlbom et al., 2003a, b; Jin et al., 2005;
Karhumaa et al., 2005; Pitkänen et al., 2005; Parachin
et al., 2010). Strain CMB.GS010 clearly exhibited a respiratory metabolism on this sugar. Xylose utilization was
almost entirely oxidative as indicated by the respiratory
coefficient (RQ = CER/OUR), which remains close to 1
during the entire cultivation time (data not shown), and
the high carbon fraction of xylose converted to biomass
as compared to glucose metabolism. Furthermore, the
physiological observations were supported by transcriptome data analysis at global and metabolic level. The
most over-represented gene families in the evolved strain
were related to functions or features linked to respiratory
process. Consistently, TFs enrichment analysis identified
factors primarily involved in carbon catabolite repression
response mechanism and regulation of the respiration.
Mainly, the significantly enriched TFs in the evolved
strain represent transcriptional activator of gene involved
in non-fermentative metabolism (Sculler, 2003). Among
them is SNF1 that is a major regulator of carbon metabolism together with several related TFs known to be
involved in the generation of precursors of energy and
linked to the activation of peroxisomal proteins (PIP2,
OAF1) (Karpichev & Small, 1998) and in the metabolism
of non-fermentable carbon sources (CAT8, MIG) (Usaite
et al., 2009). Recent transcriptome studies on recombinant S. cerevisiae strain engineered for xylose consumpFEMS Yeast Res && (2012) 1–16
11
tion with the oxoreductive pathway (XR, XDH and XK)
indicated the role of partial repression of xylose on TCA
and glyoxylate cycle (Salusjärvi et al., 2008), and the
physiology of the strain employed during this study differ
substantially from the evolved mutant reported here,
showing that on batch cultivation, the xylose consumed
was partially fermented to ethanol and acetate beside high
xylitol production. In contrast, a xylose consuming
mutant carrying the oxoreductive pathway with a
mutated XR and additionally engineered on the PPP
pathway exhibits a full physiological respiratory response
without ethanol or xylitol overflow metabolites formation
during xylose batch cultivation that correlated with the
up-regulation of the TCA cycles at the transcriptional
level (Runquist et al., 2009), which is consistent with our
findings. Moreover, physiological characterization under
continuous cultivation conditions of mutagenesis isolated
strains capable of fast growth on xylose shown a clear
Crabtree-negative characteristics (Souto-Maior et al.,
2009). the observed up-regulation of the glyoxylate pathway in the evolved strain grown on xylose compared to
growth on glucose, or the un-evolved strain grown on
glucose is in line with observations made at low dilution
rates in glucose-limited chemostat cultures in wild-type
S. cerevisiae (Regenberg et al., 2006). As an extension of
the glyoxylate pathway, IDP2 and IDP3 were up-regulated
significantly in the evolved strain grown on xylose. Xylose
metabolism requires the pentose phosphate pathway
(PPP), and the first step of the PPP involves the conversion of glucose-6-phosphate to 6-phosphogluconate,
catalysed by glucose-6-phosphate dehydrogenase (ZWF1).
The PPP is essential for the generation of biomass precursors and NADPH cofactor for anabolic reactions (Jeffries,
2006) While the non-oxidative PPP satisfies biomass precursor demands, cytosolic NADPH must still be generated, and the oxidative part of the pathway is bypassed
during growth on xylose. Cytosolic isocitrate dehydrogenase (Idp2) catalyses the oxidation of isocitrate to a-ketoglutarate and is NADP+-specific (Cherry et al., 1997). On
both fermentable and non-fermentable carbon sources,
Zwf1p is constitutively expressed while Idp2p levels are
glucose-repressed (Thomas et al., 1991; Minard et al.,
1998), whereas Idp2p levels have been demonstrated to
be elevated on non-fermentable carbon sources and during the diauxic shift as glucose is depleted (Loftus et al.,
1994; DeRisi et al., 1997; Minard et al., 1998). Furthermore, in Dzwf1 Dadh6 S. cerevisiae mutants, it was demonstrated that Idp2 is up-regulated and generates enough
NADPH to satisfy biomass requirements, noting that the
NADP+-specific cytosolic aldehyde dehydrogenase
(Adh6p) catalysing acetaldehyde conversion to acetate is
the other major cytosolic source of NADPH (Minard &
McAlister-Henn, 2005). In the evolved strain, IDP2 and
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
12
IDP3 likely provide a source of NADPH to satisfy biomass requirements.
The native xylose-fermenting strain P. stipitis, which is
the source of the heterologous expressed enzymes, XR
and XDH, does not produce xylitol during xylose fermentations (Skoog & Hahn-Hägerdal, 1990). Extensive xylitol
formation has been observed in all the S. cerevisiae xylose
consuming strains expressing these enzymes (Kötter &
Ciriacy, 1993; Tantirungkij et al., 1993; Walfridsson et al.,
1995; Ho et al., 1998; Eliasson et al., 2000; Tiovari et al.,
2001). The production of xylitol has been shown to be
the direct result of a redox imbalance of the NAD(P) cofactors between the XR and XDH (Eliasson et al., 2000;
Wahlbom & Hahn-Hägerdal, 2002; Jeppsson et al., 2003;
Roca et al., 2003; Träff-Bjerre et al., 2003; Verho et al.,
2003; Watanabe et al., 2007). This imbalance has recently
been successfully avoided by direct conversion of xylose
to xylulose via the introduction of a bacterial isomerase
(Kuyper et al., 2003, 2004). Xylitol formation is often
described as being the major drawback of the XR-XDH
strategy; however, in the engineered strain selected in this
study, the formation of xylitol was completely absent during all the xylose fermentations.
The absence of xylitol accumulation under oxidative
conditions may be interpreted as a result of complete
xylitol oxidation. Consistent with this assumption is that
reduction of xylose to xylitol by XR is limited by the
availability of NADPH. Perhaps, as the data in this study
suggest, up-regulation of IPD2 ensures sufficient NADPH
production necessary for the P. stipitis NADPH-preferring
XR to drive xylose catabolism, whereas the NADH
surplus produced is reduced through the respiration eliminating the NADP+/NAD+ imbalance. Thus, the xylose
consumption relies on the external NADH dehydrogenases
(Ned1p or Ned2p) or the mitochondrial glycerol-3-phosphate/dihydroxyacetone phosphate shuttle (Gut2p and
Gpd1p or Gpd2p) for the re-oxidation of the excess cytosolic NADH (Luttik et al., 1998; Overkamp et al., 2000).
This hypothesis is consistent with the previously reported
effect of aeration on the reduction of xylitol formation
during xylose fermentation (Winkelhausen & Kuzmanova,
1998) and could explain the inability of the evolved strain
to consume xylose under anaerobic conditions.
Co-utilization of both sugars, glucose and xylose, is
essential for an economically feasible conversion of lignocellulose to industrially relevant bioproducts. Xylose is
predominantly consumed after glucose exhaustion. This
could be explained a competitive inhibition model for the
uptake of the two sugars. Until now, no transporters have
been found in S. cerevisiae that can exclusively and specifically transport xylose. Nevertheless, it is known that
xylose competes with glucose for the same transporters
albeit their affinity for xylose is lower (Kötter & Ciriacy,
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
G. Scalcinati et al.
1993; van Zyl et al., 1999), and so xylose uptake proceeds
slower compared to glucose (Leandro et al., 2009). Sugar
uptake rate has been related to the carbon catabolite
repression and has a role in determining the control the
switch between fermentative vs. respirative metabolism
(Goffrini et al., 2001; Elbing et al., 2004 Daphne et al.,
2012). The evolved mutant displayed sensitivity to different extracellular xylose concentration. In xylose consuming mutant, sugar transport constitutes an important step
in determining the fermentation performances (Saloheimo et al., 2007; Jojima et al., 2010). However, the evolved
mutant retains a fermentative ability towards glucose
comparable to the un-evolved that prevail during mixed
sugar cultivation suggesting that the metabolic response
of the evolved strain was exclusive to xylose. During glucose–xylose mixed cultivation, the observed poor xylose
consumption in the presence of glucose has been attributed to the glucose repression effect (Belinchon &
Gancedo, 2003). Recently, several studies propose elegant
approaches to overcome the sequential utilization of glucose and xylose, acting on bypass the glucose repression
effect on xylose uptake and allowing the co-fermentation
of the two sugars (Nakamura et al., 2008; Ha et al.,
2010).
Although the efficiency of the evolutionary approach
presented in this work is promising, the underlying
genetic change that has likely taken place during the
direct evolution process remains unclear. Further studies
are necessary to gain insight into the possible mutations
that contributes to the observed physiology. A detailed
genome-wide investigation would offer the opportunity
to investigate the genetic basis that result in the ability of
the selected strain to consume efficiently xylose as sole
carbon source as demonstrated recently in a study on
galactose metabolism (Hong et al., 2011).
Acknowledgements
J.M.O. is a Merck Doctoral Fellow and acknowledges
financial support from the Division of Bioprocess
Research & Development, Merck Research Labs, Merck &
Co., Inc. Part of this research has been financed by the
Chalmers Foundation and the Knut and Alice Wallenberg
Foundation.
Authors’ contributions
G.S. and J.M.O. contributed equally to this research.
References
Amore R, Kötter P, Kuster C, Ciriacy M & Hollenberg CP
(1991) Cloning and expression in Saccharomyces cerevisiae of
FEMS Yeast Res && (2012) 1–16
Metabolic enginnering of S. cerevisiae for xylose consumption
the NAD(P)H-dependent xylose reductase-encoding gene
(XYL1) from the xylose-assimilating yeast Pichia stipitis.
Gene 109: 89–97.
Ball CA, Dolinski K, Dwight SS et al. (2000) Integrating
functional genomic information into Saccharomyces genome
database. Nucleic Acids Res 28: 77–80.
Ball CA, Jin H, Sherlock G et al. (2001) Saccharomyces genome
database provides tools to survey gene expression and
functional analysis data. Nucleic Acids Res 29: 80–81.
Belinchon MM & Gancedo JM (2003) Xylose and some nonsugar carbon sources cause catabolite repression in
Saccaromyces cerevisiae. Arch Microbiol 180: 293–297.
Boer VM, de Winde JH, Pronk JT & Piper MD (2003) The
genome-wide transcriptional responses of Saccharomyces
cerevisiae grown on glucose in aerobic chemostat cultures
limited for carbon, nitrogen, phosphorus, or sulfur. J Biol
Chem 8: 3265–3274.
Boles E, Lehnert W & Zimmermann FK (1993) The role of
NAD-dependent glutamate dehydrogenase in restoring
growth on glucose of Saccharomyces cerevisiae
phosphoglucose isomerase mutant. Eur J Biochem 217: 469–
477.
Cai Z, Zhang B & Li Y (2012) Engineering Saccharomyces
cerevisiae for efficient anaerobic xylose fermentation:
reflection and perspectives. Biotechnol J 7: 34–46.
Chemler JA, Yan Y & Koffas MA (2006) Biosynthesis of
isoprenoids, polyunsaturated fatty acids and flavonoids in
Saccharomyces cerevisiae. Microb Cell Fact 6: 20.
Cherry JM, Ball C, Weng S et al. (1997) Genetic and physical
maps of Saccharomyces cerevisiae. Nature 387: 67–73.
Christensen LH, Schulze U, Nielsen J & Villadsen J (1995)
Acoustic off gas analyzer for bioreactors: precision, accuracy
and dynamics of detection. Chem Eng Sci 50: 2601–2610.
Chu BC & Lee H (2007) Genetic improvement of
Saccharomyces cerevisie for xylose fermentation. Biotechnol
Adv 25: 425–441.
Daphne HE, Huberts DH, Neibel B & Heinemann M (2012)
A flux-sensing mechanism could regulate the switch
between respiration and fermentation. FEMS Yeast Res 12:
118–128.
DeRisi JL, Iyer VR & Brown PO (1997) Exploring the
metabolic and genetic control of gene expression on
genomic scale. Science 278: 680–686.
du Preez JC (1994) Process parameters and environmental
factors affecting D-xylose fermentation by yeast. Enzyme
Microbial Technol 16: 944–956.
Elbing K, Larsson C, Bill RM, Alberts E, Snoep JL, Boles E,
Hohmann S & Gustafsson L (2004) Role of hexose transport
in control of glycolytic flux in Saccharomyces cerevisiae. Appl
Environ Microbiol 70: 5323–5330.
Eliasson A, Christensson C, Wahlbom FC & Hahan-Hägerdal
B (2000) Anaerobic xylose fermentation by recombinant
Saccharomyces cerevisiae carrying XYL1, XYL2, and XKS1 in
mineral medium chemostat cultures. Appl Environ Microbiol
66: 3381–3386.
FEMS Yeast Res && (2012) 1–16
13
Ferea T, Botstein D, Brown PO & Rosenzweig F (1999)
Systematic changes in gene expression patterns following
adaptive evolution in yeast. P Natl Acad Sci USA 96: 9721–
9726.
Gietz RD & Woods RA (2002) Transformation of yeast by
lithium acetate/single-stranded carrier DNA/polyethylene
glycol method. Methods Enzymol 350: 87–96.
Goffrini P, Ferrero I & Donnini C (2001) Respiratorydependent utilization of sugars in yeasts: a determinant role
for sugar transporters. J Bacteriol 184: 427–432.
Gresham D, Desai MM, Tucker CM, Jenq HT, Ward A,
DeSevo CG, Botstein D & Dunham MJ (2008) The
repertoire and dynamics of evolutionary adaptation to
controlled nutrient-limited environments yeast. PLoS Genet
4: e1000303.
Ha SJ, Gakazka JM, Kim RS, Choi JH, Yang X, Seo JH, Glass
NL, Cate JHD & Jin YS (2010) Engineered Saccharomyces
cerevisiae capable of simultaneous cellobiose and xylose
fermentation. P Natl Acad Sci USA 108: 504–509.
Hahn-Hägerdal B, Karhumaa K, Jeppson M & GorwaGrausland MF (2007) Metabolic engineering for pentose
utilization in Saccharomyces cerevisiae. Adv Biochem Eng
Biotechnol 108: 147–177.
Haiying N, Laplaza JM & Jeffries TW (2007) Transposon
mutagenesis to improve the growth of recombinant
Saccharomyces cerevisiae on D-xylose. Appl Environ Microbiol
73: 2061–2066.
Harhangi HR, Akhmanova AS, Emmens R, van der Drift C, de
Laat WT, van Dijken PJ, Jetten MS, Pronk JT & Op den
Camp HJ (2003) Xylose metabolism in the fungus Piromyces
sp. strain E2 follows the bacterial pathway. Arch Microbiol
180: 134–141.
Herrero PJ, Galindez N, Ruiz C, Martinez-Campa &
Moreno F (1995) Transcriptional regulation of the
Saccharomyces cerevisiae HXK1, HXK2 and GLK1 genes.
Yeast 11: 137–144.
Ho NW, Chen Z & Brainard AP (1998) Genetically engineered
Saccharomyces yeast capable of effective co-fermentation of
glucose xylose. Appl Environ Microbiol 64: 1852–1859.
Hong K, Vongsangnak W, Vemuri GN & Nielsen J (2011)
Unravelling evolutionary strategies of yeast for improving
galactose utilization through integrated systems level
analysis. P Natl Acad Sci USA 108: 12179–12184.
Jansen LA, Diderich AJ, Mashego A, Hassane A, de Winde HJ,
Daran-Lapujade P & Pronk J (2005) Prolonged selection in
aerobic, glucose-limited chemostat cultures of Saccharomyces
cerevisiae causes a partial loss of glycolitic capacity.
Microbiology 151: 1657–1669.
Jeffries TW (2006) Engineering yeasts for xylose metabolism.
Curr Opin Biotechnol 17: 320–326.
Jeppsson M, Träff-Bjerre KL, Johansson B, Hahn-Hägerdal B
& Gorwa-Grauslund MF (2003) Effect of enhanced xylose
reductase activity on xylose consumption and product
distribution in xylose-fermenting recombinant
Saccharomyces cerevisiae. FEMS Yeast Res 3: 167–175.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
14
Jin YS, Lapaza JM & Jeffries TW (2004) Saccharomyces
cerevisiae engineered for xylose metabolism exhibits a
respiratory response. Appl Environ Microbiol 70: 6816–6825.
Jin YS, Alper H, Yang YT & Stephanopolous G (2005)
Improvement of xylose uptake and ethanol production in
recombinant Saccharomyces cerevisiae through an inverse
metabolic engineering approach. Appl Environ Microbiol 71:
8249–8256.
Jojima T, Omumasaba CA, Inui M & Yukawa H (2010) Sugar
transporters in efficient utilization of mixed sugar
substrates: current knowledge and outlook. Appl Microbiol
Biotechnol 85: 417–480.
Karhumaa K, Hahn-Hägerdal B & Gorwa-Grausland MF
(2005) Investigation of limiting metabolic steps in the
utilization of xylose by recombinant Saccharomyces cerevisiae
using metabolic engineering. Yeast 22: 359–368.
Karpichev IV & Small GM (1998) Global regulatory functions
of Oaf1p and Pip2 (Oaf2p), transcription factors that
regulate genes encoding peroxisomal proteins in
Saccharomyces cerevisiae. Mol Cell Biol 18: 6560–6570.
Kim IK, Roldão A, Siewers V & Nielsen J (2012) A systemslevel approach for metabolic engineering of yeast cell
factories. FEMS Yeast Res 12: 228–248.
Kötter P & Ciriacy M (1993) Xylose fermentation by
Saccharomyces cerevisiae. Appl Microbiol Biotechnol 38: 776–
783.
Kötter P, Amore R, Hollenberg CP & Ciriacy M (1990)
Isolation and characterization of Pichia stipitis xylitol
dehydrogenase gene, XYL2, and construction of a xyloseutilizing Saccharomyces cerevisiae transformant. Curr Genet
18: 493–500.
Kumar D & Murthy G (2011) Impact of pretreatment and
downstream processing technologies on economics and
energy in cellulosic ethanol production. Biotechnol Biofuels
4: 27–46.
Kuyper M, Harhangi HR, Stave AK, Winkler AA, Jetten MS,
de Laat WT, den Ridder JJ, Op den Camp HJ, van Dijken
JP & Pronk JT (2003) High-level functional expression of
fungal xylose isomerase: the key to efficient ethanolic
fermentation of xylose by Saccharomyces cerevisiae? FEMS
Yeast Res 4: 69–78.
Kuyper M, Winkler AA, van Dijken JP & Pronk JT (2004)
Minimal metabolic engineering of Saccharomyces cerevisiae
for efficient anaerobic xylose fermentation: proof of
principle. FEMS Yeast Res 4: 655–664.
Kuyper M, Toirkens MJ, Diderich JA, Winkler AA, van Dijken
JP & Pronk JT (2005) Evolutionary engineering of mixedsugar utilization by a xylose-fermenting Saccharomyces
cerevisiae strain. FEMS Yeast Res 5: 925–934.
Leandro MJ, Fonseca C & Goncalves P (2009) Hexose and
pentose transport in ascomycetus yeasts: an overview. FEMS
Yeast Res 9: 511–525.
Loftus TM, Hall LV, Anderson SL & McAlister-Henn L (1994)
Isolation, characterization, and disruption of the yeast gene
encoding cytosolic NADP-specific isocitrate dehydrogenase.
Biochemistry 33: 9661–9667.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
G. Scalcinati et al.
Luttik MA, Overkamp KM, Kötter P, de Vries S, van Dijken
JP & Pronk JT (1998) The Saccharomyces cerevisiae NDE1
and NDE2 genes encode separate mitochondrial NADH
dehydrogenases catalyzing the oxidation of cytosolic NADH.
J Biol Chem 273: 24529–24534.
Matsushika A, Inoue H, Kodaki T & Sawayama S (2009)
Ethanol production from xylose in engineered
Saccharomyces cerevisiae strains: current state and
perspectives. Appl Microbiol Biotechnol 84: 37–53.
Minard KI & McAlister-Henn L (2005) Sources of NADPH in
yeast vary with carbon source. J Biol Chem 280: 39890–
39896.
Minard KI, Jennings GT, Loftus TM, Xuan D & McAlisterHenn L (1998) Sources of NADPH and expression of
mammalian NADP+-specific isocitrate dehydrogenases
in Saccharomyces cerevisiae. J Biol Chem 273: 31486–
31493.
Misha P & Singh A (1993) Microbial pentose utilization. Adv
Appl Microbiol 39: 91–152.
Müller S, Boles E, May M & Zimmermann FK (1995)
Different internal metabolites trigger the induction of
glycolytic gene expression in Saccharomyces cerevisiae. J
Bacteriol 177: 4517–4519.
Nakamura N, Yamada R, Katahira S, Tanaka T, Fukuda H &
Kondo A (2008) Effective xylose/cellobiose co-fermentation
and ethanol production by xylose-assimilating S. cerevisiae
via expression of ß-glucosidase on its cell surface. Enzyme
Microb Technol 43: 233–236.
Nielsen J & Olsson L (1997) On-Line in situ monitoring of
biomass in submerged cultivations. Trends Biotechnol 15:
517–522.
Olsson L, Jørgensen H, Krogh K & Roca C (2004). Bioethanol
production from lignocellulosic material. Polysaccharides:
Structural Diversity and Functional Versatility (Dumitriu S,
ed), pp. 957–993. Dekker M Inc., New York.
Ostergaard S, Olsson L & Nielsen J (2000) Metabolic
engineering of Saccharomyces cerevisiae. Microbiol Mol Biol
Rev 64: 34–50.
Otero JM, Panagiotou G & Olsson L (2007) Fueling industrial
biotechnology growth with bioethanol. Adv Biochem Eng
Biotechnol 108: 1–40.
Overkamp KM, Bakker BM, Kötter P, van Tuijl A, de Vries S,
van Dijken JP & Pronk JT (2000) In vivo analysis of the
mechanisms for oxidation of cytosolic NADH by
Saccharomyces cerevisiae mitochondria. J Bacteriol 10: 2823–
2830.
Parachin NS, Bengtsson O, Hahn-Hägerdal B & GorwaGrausland MF (2010) The deletion of YLR042c improves
ethanolic xylose fermentation by recombinant Saccharomyces
cerevisiae. Yeast 27: 741–751.
Patil KR & Nielsen J (2005) Uncovering transcriptional
regulation of metabolism by using metabolic network
topology. P Natl Acad Sci USA 102: 2685–2689.
Pitkänen JP, Rintala E, Aristidou A, Ruohonen L & Penttilä M
(2005) Xylose chemostat isolates of Saccharomyces cerevisiae
show altered metabolite and enzyme levels compared with
FEMS Yeast Res && (2012) 1–16
Metabolic enginnering of S. cerevisiae for xylose consumption
xylose, glucose, and ethanol metabolism of the original
strain. Appl Microbiol Biotechnol 67: 827–837.
Regenberg B, Groktjær T, Winther O, Fausbøll A, Åkesson M,
Bro C, Hansen LK, Brunak S & Nielsen J (2006) Growthrate regulated genes have profound impact on interpretation
of transcriptome profiling in Saccharomyces cerevisiae.
Genome Biol 7: R107.
Roca C, Nielsen J & Olsson L (2003) Metabolic engineering of
ammonium assimilation in xylose-fermenting Saccharomyces
cerevisiae improves ethanol production. Appl Environ
Microbiol 69: 4732–4736.
Runquist D, Hahn-Hägerdal B & Bettiga M (2009) Increased
expression of the oxidative pentose phosphate pathway and
gluconeogenesis in anaerobically growing xylose-utilizing
Saccharomyces cerevisiae. Microb Cell Fact 8: 49.
Saloheimo A, Rauta J, Stasyk OV, Sibirny AA, Penttilä M &
Ruohonen L (2007) Xylose transport studies with xyloseutilizing Saccharomyces cerevisiae strains expressing
heterologous and homologous premeases. Appl Micorbiol
Biotechnol 74: 1041–1052.
Salusjärvi L, Pitkänen JP, Aristidou A, Ruohonen L & Penttilä
(2006) Transcription analysis of recombinant Saccharomyces
cerevisiae reveals novel response to xylose. Appl Biochem
Biotechnol 128: 237–261.
Salusjärvi L, Kankainen M, Soliymani R, Pitkänen JP, Penttilä M
& Ruohonen L (2008) Regulation of xylose metabolism in
recombinant Saccharomyces cerevisiae. Microb Cell Fact 7: 18.
Sculler HJ (2003) Transcriptional control of nonfermentative
metabolism in yeast Saccharomyces cerevisiae. Curr Genet 43:
139–160.
Senac T & Hahn-Hägerdal B (1991) Effects of increased
transaldolase activity on D-xylulose and D-glucose
metabolism in Saccharomyces cerevisiae cell extracts. Appl
Environ Microbiol 57: 1701–1706.
Skoog K & Hahn-Hägerdal B (1990) Effect of oxygenation on
xylose fermentation by Pichia stipitis. Appl Environ Microbiol
56: 3389–3394.
Smyth GK (2005) Limma: linear models for microarray data.
Bioinformatics and Computational Biology Solutions using R
and Bioconductor (Gentleman R, Carey V, Dudoit S,
Irizarry R & Huber W, eds), pp. 397–420. Springer, New
York.
Sonderegger M & Sauer U (2003) Evolutionary engineering of
Saccharomyces cerevisiae for anaerobic growth on xylose.
Appl Environ Microbiol 69: 1990–1998.
Souto-Maior AM, Runquist D & Hahn-Hägerdal B (2009)
Crabtree-negative characteristics of recombinant xyloseutilizing Saccharomyces cerevisiae. J Biotechnol 143: 119–123.
Stephanopoulos G (2010) Challenge in engineering microbes
for biofuels production. Science 315: 801–804.
Tantirungkij M, Nakashima N, Seki T & Yoshida T (1993)
Construction of xylose-assimilating Saccharomyces cerevisiae.
J Ferment Bioeng 75: 83–88.
Thomas D, Cherest H & Surdin-Kerjan Y (1991) Identification
of the structural gene for glucose-6-phosphate
dehydrogenase in yeast. EMBO J 10: 547–553.
FEMS Yeast Res && (2012) 1–16
15
Tiovari HM, Aristidou A, Ruohonen L & Penttilä M (2001)
Conversion of xylose to ethanol by recombinant
Saccharomyces cerevisiae: importance of xylulokinase (XKS1)
and oxygen availability. Metab Eng 3: 236–249.
Träff-Bjerre KL, Jeppsson M, Hahn-Hägerdal B & GorwaGrauslund MF (2003) Endogenous NADPH-dependent
aldose reductase activity influences products formation
during xylose consumption in recombinant Saccharomyces
cerevisiae. Yeast 21: 141–150.
Tyo KE, Fisher CR, Simeon F & Stephanopoulos G (2010)
Analysis of polyhydroxybutyrate flux limitations by
systematic genetic and metabolic perturbations. Metab Eng
12: 187–195.
Usaite R, Jewett MC, Oliveira AP, Yates JR 3rd, Olsson L &
Nielsen J (2009) Reconstruction of the yeast Snf1 kinase
regulatory network reveals its roles as global energy
regulator. Mol Syst Biol 5: 319.
van Dijken JP, Bauer J, Brambilla L et al. (2000) An
interlaboratory comparison of physiological and genetic
proprieties of four Saccharomyces cerevisiae strains. Enzyme
Microb Technol 26: 706–741.
van Maris AJ, Winkler AA, Kuyper M, de Laat WT, van
Dijken JP & Pronk JT (2007) Development of efficient
xylose fermentation in Saccharomyces cerevisiae: xylose
isomerase as a key component. Adv Biochem Eng Biotechnol
108: 179–204.
Van Vleet JH & Jeffries TW (2009) Yeast metabolic
engineering for hemicellulosic ethanol production. Curr
Opin Biotechnol 20: 300–306.
van Zyl WH, Eliasson A, Hobley T & Hahn-Hägerdal B (1999)
Xylose utilization by recombinant strains of Saccharomyces
cerevisiae on different carbon source. Appl Microbiol
Biotechnol 52: 829–833.
Verduyn C, Postma E, Scheffers A & van Dijken P (1992)
Effect of benzoic acid on metabolic fluxes in yeasts: a
continuous culture study on the regulation of respiration
and alcoholic fermentation. Yeast 8: 501–517.
Verho R, Londesborough J, Penttilä M & Richard P (2003)
Engineering redox cofactor regeneration for improved
pentose fermentation in Saccharomyces cerevisiae. Appl
Environ Microbiol 69: 4732–4736.
Wahlbom CF & Hahn-Hägerdal B (2002) Furfural, 5hydroxymethyl furfural, and acetoin act as external electron
acceptors during anaerobic fermentation of xylose in
recombinant Saccharomyces cerevisiae. Biotechnol Bioeng 78:
172–178.
Wahlbom CF, Cordero Otero RR, van Zyl WH, HahnHägerdal B & Jönsson LJ (2003a) Molecular analysis of a
Saccharomyces cerevisiae mutant with improved ability to
utilize xylose shows enhanced expression of proteins
involved in transport, initial xylose metabolism, and the
pentose phosphate pathway. Appl Environ Microbiol 69:
740–746.
Wahlbom CF, van Zyl WH, Jönsson LJ, Hahn-Hägerdal B &
Otero RR (2003b) Generation of the improved recombinant
xylose-utilizing Saccharomyces cerevisiae TMB 3400 by
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
16
random mutagenesis and physiological comparison with
Pichia stipitis CBS 6054. FEMS Yeast Res 3: 319–326.
Walfridsson M, Hallborn J, Penttilä M, Keränen S & HahnHägerdal B (1995) Xylose-metabolizing Saccharomyces
cerevisiae strains overexpressing TKL1 and TAL1 genes
encoding the pentose phosphate pathway enzymes
transketolase and transaldolase. Appl Environ Microbiol 61:
4184–4190.
Wang PY, Shopsis C & Schneider H (1980) Fermentation of
pentose by yeasts. Biochem Biophys Res Commun 94: 248–254.
Watanabe S, Saleh AA, Pack SP, Annaluru N, Kodaki T &
Makino K (2007) Ethanol production from xylose by
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
G. Scalcinati et al.
recombinant Saccharomyces cerevisiae expressing protein
engineered NADP+-dependent xylitol dheydrogenase. J
Biotechnol 130: 316–319.
Winkelhausen E & Kuzmanova S (1998) Microbial conversion
of D-xylose to xylitol. J Ferment Bioeng 86: 1–14.
Yamano S, Ishii T, Nakagawa M, Ikenaga H & Misawa N
(1994) Metabolic engineering for production of ß-carotene
and lycopene in Saccharomyces cerevisiae. Biosci Biotechnol
Biochem 58: 1112–1114.
FEMS Yeast Res && (2012) 1–16